Novels

Controlling the conjugation length in poly[5- n-butyl-2-(2-ethylhexyl)-1,4-phenylenevinylene]: exploring the scope of hydrogen radical substitution of leaving groups on precursor polymers

Description
Controlling the conjugation length in poly[5- n-butyl-2-(2-ethylhexyl)-1,4-phenylenevinylene]: exploring the scope of hydrogen radical substitution of leaving groups on precursor polymers
Categories
Published
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Synthetic Metals 145 (2004) 159–169 Controlling the conjugation length inpoly[5- n -butyl-2-(2-ethylhexyl)-1,4-phenylenevinylene]:exploring the scope of hydrogen radical substitutionof leaving groups on precursor polymers G.R. Webster, P.L. Burn ∗  Department of Chemistry, Chemistry Research Laboratory, University of Oxford, Mansfield Road, Oxford OX1 3TA, UK  Received 22 March 2004; received in revised form 22 April 2004; accepted 26 April 2004Available online 28 July 2004 Abstract Precursorroutestopoly(arylenevinylene)sprovideapowerfulmethodforforminginsolublefullydensethinfilmsofconjugatedpolymers.The photophysical and device properties of these conjugated polymer films can be altered by the introduction of saturated units betweenthe conjugated segments. We have found that tri- n -butyltin hydride can be used to substitute chloride, bromide, or  O -ethylxanthate leavinggroups on the precursor polymer backbone with hydrogen to give ethylene units. The scope of the technique was demonstrated on thethree different precursor polymers to poly[5- n -butyl-2-(2-ethylhexyl)-1,4-phenylenevinylene]. The degree of substitution of each leavinggroup was found to be easily controlled by the number of equivalents of tri- n -butyltin hydride used and the yields of the hydrogen radicalsubstitutedpolymerswereintherangeof50–91%.Whereaproportionoftheleavinggroupswereremoved,theprecursorcopolymercouldbethermallyconvertedtothecorrespondingconjugatedpolymerthatcontainedthesamenumberofsaturatedlinkages.The O -ethylxanthateprecursor polymer was found to be essentially free of conjugated segments both after polymerisation and the hydrogen radical substitution.Both the chloride and bromide precursors had low levels of conjugation after polymerisation with the amount increasing for the bromideduring the substitution reaction.© 2004 Elsevier B.V. All rights reserved. Keywords:  Poly[5- n -butyl-2-(2-ethylhexyl)-1,4-phenylenevinylene]; Hydrogen radical substitution; Polymers; PLEDs and morphology 1. Introduction Poly(fluorene) (PF) and poly(1,4-phenylenevinylene)(PPV) based materials comprise the two main fami-lies of light-emitting polymers that are used in polymerlight-emitting diodes (PLEDs) [1]. For both the families, the polymer morphology plays a critical role in controlling thephotophysical and device properties of the materials [2–6].PFs require lipophilic side groups to be attached to ensuresolubility in solvents suitable for processing into thin films.In contrast, PPVs can be formed with solubilising lipophilicgroups or via precursor polymers. The use of the precur-sor route allows the direct formation of fully dense filmsof low solubility [5,7]. Another property of the precursorroute is that it provides a pathway for the incorporation ∗ Corresponding author. Fax:  + 44 1865 285002.  E-mail address:  paul.burn@chem.ox.ac.uk (P.L. Burn). of saturated units into the conjugated polymer backbone[8–18]. The introduction of saturated units into the poly-mer backbone not only effects the conjugation length butalso the morphology of the polymer. This has led to thediscovery that in some cases the presence of small levels of saturation in the polymer backbone can improve the emis-sive properties of PLEDs based on PPVs [8–11,16,19]. In addition, the controlled introduction of saturated units intothe PPV backbone has proved a useful tool for studyingthe photophysical and charge mobility properties of PPVbased polymers [12–15,18,20–22]. However, although theemissive properties of the materials might improve withthe introduction of saturated units it has been shown thatacetylene and ethylene defects in the conjugated polymerbackbone, introduced during the synthesis, can lead to de-vice instability [23]. What is not clear from this latter studyis which of the defects, acetylene or ethylene, leads to de-vice instability. Given that alkyl groups are used as sidechains on soluble conjugated polymers it would seem more 0379-6779/$ – see front matter © 2004 Elsevier B.V. All rights reserved.doi:10.1016/j.synthmet.2004.04.019  160  G.R. Webster, P.L. Burn/Synthetic Metals 145 (2004) 159–169 likely that the acetylene units play the main role in reducingdevice lifetime. As the acetylene units arise from ethyleneunits in the precursor polymer that have two leaving groupsattached it would be adventitious to have a method that re-duces their level without having to resort to major changesto the polymer structure.There are only a few methods described for the intro-duction of non-conjugated sequences into the backbone of PPV derivatives prepared via precursor polymers. The mainstrategy has been to have a precursor polymer with a com-bination of two different leaving groups with one being lesslabile to the conversion conditions [8–15]. Thus, on conver-sion the less labile moiety remains on the polymer backbonewith the more labile leaving group being eliminated to giveconjugated sequences. Both acetoxy and alkoxy groups havebeen used as the less labile leaving group [8–15]. Partialelimination of a single leaving group has also being usedto form partially conjugated PPVs [16,17]. More recentlyan interesting approach to partial conjugation in PPVs hasbeenreportedwherebysaturatedunitsareintroducedintothepolymer backbone by reaction of a proportion of the viny-lene units in a fully conjugated PPV [18]. These strategies all have a potential weakness in that the less labile leavinggroups can eliminate slowly over time or even during mea-surements designed to ascertain the properties of the mate-rials.In this paper, we describe an alternative approach to par-tially conjugated PPVs, which is to remove a proportion of the leaving groups by hydrogen radical substitution to give aprecursor polymer with unsubstituted ethylene and leavinggroup substituted ethylene linkages. The precursor copoly-mer thus formed can then be converted to the correspondingpartially conjugated copolymer containing a proportion of ethylene units in the backbone without leaving groups at-tached. Such materials are clearly of interest for the study of the effect of morphology on the photophysical and chargemobility properties. In addition, the process should also re-duce the level of acetylene defects in the backbone whilstintroducing controlled levels of ethylene units. Therefore,the materials could also be useful models for understandingthe role of defect type on device lifetime.Halo [24], sulfonium [25], xanthate [26], and sulfoxide [27] precursor routes are all used for the preparation of PPV derivatives. Halo and xanthate groups can be sub-stituted with hydrogen radicals in small molecules andhere we explore the reaction in the context of leavinggroup precursor polymers. The polymer chosen for thisstudy was the highly fluorescent dialkyl substituted PPVpoly[2- n -butyl-5-(2-ethylhexyl)-1,4-phenylenevinylene](BuEHPPV) [28]. We describe for the first time the syn- thesis of BuEHPPV via three different precursor routes,namely, chloride, bromide and  O -ethylxanthate. We alsodemonstrate that the hydrogen radical reaction is an excel-lent method for forming partially conjugated dialkyl-PPVsand is compatible with chloride, bromide and xanthateleaving groups. Scheme 1. (a) 2-Ethylhexanoyl chloride, AlCl 3 ; (b) Zn, HgCl 2 , MeOH,H 2 O, HCl ( aq ) ,   ; (c) 47% (w/v) HBr/AcOH, AcOH, (CH 2 O) n , ZnBr 2 ,80 ◦ C then AcOH, NaOAc, then 10% (w/v) NaOH ( aq ) , MeOH,   ; (d)SOCl 2 ; (e) PBr 3 , 100 ◦ C; (f) KSCSOEt,  n -Bu 4 NBr, DCM,   . 2. Results and discussion The synthetic pathways to the monomers and polymersare illustrated in Schemes 1 and 2. 2.1. Monomer syntheses The critical intermediate for the formation of the chlo-ride, bromide, and  O -ethylxanthate monomers was thebis(hydroxymethyl)benzene  5  (Scheme 1). The first stepto  5  was the Friedel–Crafts acylation of   n -butylbenzene 1  with 2-ethylhexanoyl chloride and aluminium trichlo-ride. Under these conditions ketone  2  was formed in a ≈ 95% yield. The  1 H NMR of   2  indicated that it containeda small amount of a second compound, which could notbe separated. As the elemental analysis for  2  was correctwe believe the second compound was probably an isomerof   2 . Ketone  2  was reduced by a modified Clemmensenreduction [29] and alkane  3  was isolated as a single iso-mer in a yield of 63%. bis-Bromomethylation of   3  withzinc bromide, paraformaldehyde and hydrogen bromide inglacial acetic acid at 80 ◦ C was found to be slow. Even afterextended reaction time mono-bromomethylated benzeneswere present in addition to the desired product  7 .  7  and themono-bromomethylated benzenes could not be separatedand so  7  was converted to the bis(hydroxymethyl)benzene 5 , via the bis(acetoxymethyl)benzene  4 . Conversion to the  G.R. Webster, P.L. Burn/Synthetic Metals 145 (2004) 159–169  161Scheme 2. (a) 0.9 equiv. KOBu t  , THF; (b) heat, vacuum; (c) ( n -Bu) 3 SnH, AIBN, THF,   . It should be noted that the copolymers are not composed of well-defined blocks but have the individual units randomly distributed along the polymer backbone. bis(hydroxymethyl)benzene  5  was achieved by heating themixture of   7  and the mono-bromomethylated benzenes withsodium acetate in acetic acid at reflux for 5h. Followingwork-up of the reaction the mixture of   4  and mono-acetateswas saponified with sodium hydroxide in a methanol–watermixture. Separation of the hydroxymethylbenzenes waspossible and bis(hydroxymethyl)benzene  5  was isolated inan overall yield of 76% from  3 .The bis(chloromethyl)benzene monomer  6  was pre-pared in an 88% yield from  5  by reaction with anexcess of thionyl chloride at room temperature. Thebis(bromomethyl)benzene monomer  7  was prepared byheating  5  with phosphorus tribromide at 100 ◦ C and isolatedin a yield of 85%. The  p -xylylene-bis( O -ethylxanthate)monomer  8  was prepared in a 99% yield from the reactionof potassium  O -ethylxanthate with  7  in dichloromethaneheated at reflux in the presence of a phase transfercatalyst. 2.2. Polymer syntheses and characterisation The three monomers were all polymerised under simi-lar conditions to give their respective precursor polymers.In each case a solution of 0.9 equivalents of potassium tert  -butoxide in tetrahydrofuran was added to a solutionof the monomer in tetrahydrofuran. The reactions werequenched and the polymers purified by precipitation intomethanol. The chloride  9 , bromide  10 , and  O -ethylxanthate 11  precursor polymers were formed in yields typical forthese type of precursor polymers and were 45, 42, and 30%,respectively. The polymers showed good solubility in polaraprotic solvents such as chloroform and tetrahydrofuran.The molecular weights of the precursor polymers werestudied by gel permeation chromatography (GPC) againstpolystyrene standards. In common with other halo precur-sor polymers the molecular weights of the chloride  9  andbromide  10  precursor polymers were strongly dependent on  162  G.R. Webster, P.L. Burn/Synthetic Metals 145 (2004) 159–169 the conditions under which the measurement was carriedout. For  9  when the polymer solution was diluted and runimmediately the  M  w  was found to be >3.5  ×  10 6 with thepolydispersity ≈ 1.2. However, after equilibration of the di-lute solution overnight or by sonication the  M  w  decreasedto 2.0  ×  10 5 . As reported for other halo precursor poly-mers the decrease  M  w  is not due polymer degradation, butis thought to be due to a change in the level of the polymeraggregation in solution to which the GPC measurement isparticularly sensitive [30,31]. The  M  w  of the bromide pre-cursor  10  after initial dilution was found to be >1.9  ×  10 6 and it had a polydispersity ≈ 2.0. After equilibration by son-ication the  M  w  was found to decrease to 1.0  ×  10 5 . Themolecular weight of the  O -ethylxanthate precursor  11  wasessentially insensitive to equilibration and its  M  w  and poly-dispersity were found to be 1.4 × 10 5 and 4.1, respectively.Conjugated polymer aggregates in solution have been shownto survive spin-coating conditions with the aggregation inthe subsequent films affecting the photophysical propertiesof the polymer [32]. Hence, the difference in aggregation of  the precursor polymers in solution could also lead to differ-ences in film morphology of the final conjugated polymer.The structures of PPV precursor polymers are oftenshown as the homo-polymer. However, the polymers canoften contain some conjugation arising from base catalysedelimination during the polymerisation. In addition, the poly-merisation mechanism can give rise to the possibility of head-to-head and tail-to-tail arrangements of the monomerunits in the polymer backbone [23].  1 H NMR analysis of theprecursor polymers showed that the polymer backbone of the chloride  9  and bromide  10  precursor polymers contained ≈ 9 and  ≈ 15% vinylene units, respectively. In contrast thelevel of vinylene units in the polymer backbone was be-low an analysable level for the  O -ethylxanthate precursorpolymer  11 . The amount of conjugation formed during thepolymerisation in the three precursor polymers follows theorder of the susceptibility of the leaving groups to undergoE2 elimination and hence the level of conjugation decreasesin order from bromide to chloride to  O -ethylxanthate.Thermogravimetric analysis (TGA) further confirmed thestructures of the precursor polymers and the levels of viny-lene units within the backbone. The TGA was carried out ata heating rate of 10 ◦ C/min. For the chloride precursor  9  theeliminationtemperature( T  e )ofhydrogenchloridewasfoundto be 246 ◦ C. The measured mass loss of 10% was close tothe 11% expected for the precursor containing around 9%vinylene units. A second weight loss at 450 ◦ C was due todecomposition of the formed BuEHPPV. TGA of the bro-mide precursor polymer  10  found that the  T  e  for hydrogenbromide was 180 ◦ C. The observed weight loss of 20% wasconsistent with the precursor polymer having ≈ 15% conju-gation. The second weight loss at 445 ◦ C was again due tothe decomposition of the conjugated polymer. Finally, theelimination of the  O -ethylxanthate group from  11  occurredat  T  e = 240 ◦ C. The measured weight loss of 32% was closeto the theoretical weight loss for the polymer with no viny-lene units in the backbone and the decomposition of the re-sultant BuEHPPV occurred at 440 ◦ C.The chloride  9 , bromide  10 , and  O -ethylxanthate  11  pre-cursors were processed into thin films and converted intoBuEHPPV with the conjugated polymer formed designatedas Cl-BuEHPPV  12 , Br-BuEHPPV  13 , and X-BuEHPPV 14 , respectively. The precursor polymers were heated athigh temperature under vacuum to facilitate the conversionto the conjugated polymer. The temperature of conversionwas in each case guided by the  T  e  determined from thecorresponding TGA trace. The conversion was followed byUV–vis and infrared spectroscopy. The infrared spectra of the formed Cl-BuEHPPV, Br-BuEHPPV, and X-BuEHPPVwere identical showing that the chemical structure of theBuEHPPV formed was independent of precursor used. Theinfrared spectra of   11  and X-BuEHPPV are shown in Fig. 1to illustrate the conversion process. The  O -ethylxanthateprecursor  11  has two strong absorptions attributable to the O -ethylxanthate leaving groups at 1048 and 1210cm − 1 .On conversion these absorptions are absent from the filmshowing that full conversion has taken place. The infraredspectra of the BuEHPPV films also confirmed the  trans conformation of the double bonds with the BuEHPPV filmshaving and absorption at around 961cm − 1 correspondingto the  trans -vinylene C–H out-of-plane bend (Fig. 1).The UV–vis spectra of Cl-BuEHPPV  12 , Br-BuEHPPV 13 , and X-BuEHPPV  14  are shown in Fig. 2. All three poly- mers have similar spectra which exhibit two main featurescommon to PPV based polymers. The short wavelength fea-ture has a peak at around 215nm that corresponds to lo-calised  –  ∗ transitions. The onset of the longer wavelengthfeature of the BuEHPPV formed from the three precursorpolymers was around 510nm with a peak near 410nm. Thislonger wavelength absorption is due to the   –  ∗ transi-tions of the delocalised conjugated backbone. The onset and Fig. 1. Infrared spectra of   11  and X-BuEHPPV  14 .  G.R. Webster, P.L. Burn/Synthetic Metals 145 (2004) 159–169  163Fig. 2. UV–vis absorption spectra of Cl-BuEHPPV  12 , Br-BuEHPPV  13 ,and X-BuEHPPV  14 . The spectra have been normalised at the delocalised  –  ∗ transition for ease of comparison. peaks of the long wavelength absorptions of the BuEHP-PVs prepared from the precursors were similar to that re-ported for the soluble form of BuEHPPV [28]. An insightinto the degree of conjugation can be gained by comparisonof the ratios of the localised to delocalised transitions withthe smaller the ratio the better the delocalisation. The factthat the absorptions for the localised and delocalised   –  ∗ transitions are of nearly equal intensity for each of the con- jugated polymers shows that they all have a good level of delocalisation. 2.3. Hydrogen radical substitution and characterisation Having prepared the three precursor polymers ( 9 ,  10 , and 11 ) with different leaving groups, the scope of the hydrogenradical substitution of the leaving groups and its applica-bility to dialkyl-PPVs could be investigated. In addition todetermining whether each of the three leaving groups couldbe substituted it was also necessary to ascertain whethercontrol over the level of substitution could be achieved.Hence, one level of partial removal, that is, 25% (denoted as“25%”), and full removal (“100%”) of each of the leavinggroups was carried out. The general method for the hydro-gen radical substitution involved the stock solution of pre-cursor polymer being diluted and ultrasonicated to decreaseaggregation levels to prevent any gelation arising from inter-molecular cross-linking. Tri- n -butyltin hydride and initiatorwere added to the dilute solution of precursor polymer andthe reaction mixture was heated at reflux. The equivalents of tri- n -butyltin hydride needed for a particular level of leavinggroup substitution was based on the molecular weight of aunit of the leaving group homo-polymer. The modified pre-cursor polymers were purified by precipitation into ethanol.The yields of the polymers with the leaving groups par-tially or fully removed were in the range of 50–91%. GPCanalysis of the polymers formed after the substitution re-action showed that the molecular weights of the polymersformed were similar to the molecular weights of the dis-aggregated starting polymers. This, in combination withthe yields, clearly indicates that polymer fragmentation andintermolecular cross-linking have not occurred. If fragmen-tation had occurred, GPC analysis would have shown sig-nificantly lower molecular weights for the polymers whilstintermolecular cross-linking occurred then the observedmolecular weights would have been higher. 1 H NMR was used to determine the level of saturatedunits as well as the effect that the substitution reaction hadon the amount of vinylene units present in the polymer back-bone. The backbone of the “25%”  O -ethylxanthate precur-sor  19  formed from  11  was found to comprise of backboneethylene ( ≈ 27%) and leaving group ethylene (LG-ethylene)( ≈ 73%) units. For the “100%” polymer  20  formed from  11 no LG-ethylenes were observed.  1 H NMR also showed thatthe conditions used for the hydrogen radical substitution didnot measurably increase the number of vinylene units in  19 and  20  when compared with  11 . For the “25%” chloride pre-cursor  15  the percentages of LG-ethylene, backbone ethy-lene, and vinylene units were found to be approximately 65,27,and8%,respectively.Finally,the“25%”bromideprecur-sor  17  was determined to have, approximately, 48, 26, and26% LG-ethylene, backbone ethylene, and vinylene units,respectively. Therefore, for all three “25%” polymers,  15 , 17 , and  19 , the level of ethylene units was close to the ex-pected 25%. These results indicate that good control over thehydrogen radical substitution of the three different leavinggroups in conjunction with a dialkyl-PPV precursor can beachieved. In addition, it is important to note that the level of vinylene units was unchanged for the “25%” chloride pre-cursor polymer  15  when compared to the starting chlorideprecursor polymer  9  showing that the chloride leaving groupwas stable to thermal elimination during the radical reaction.In contrast, the “25%” bromide polymer  17  had a higherlevel of vinylene units than the starting  10  consistent with itbeing more thermally labile. For both the “100%” polymers 16  and  18  formed from  9  and  10 , respectively, the absence of the methine protons attached to the leaving groups in the  1 HNMR showed that all the leaving groups had been removed.For the “100%” polymer  16  formed from the chloride pre-cursor polymer  9  the level of vinylene units was again foundnot to increase whilst for the “100%” polymer  18  formedfrom the bromide precursor  10  the level of conjugation wasseen to increase from ≈ 15% in  10  to ≈ 24% in  18 .Further confirmation for the control over the substitutionreaction came from TGA. For the “25%” chloride precursorpolymer  15  the observed and expected weight losses were7 and 8%, respectively. TGA of “25” bromide polymer  17 gave a weight loss was 10% with the theoretical weight lossbeing 11%. Finally, the expected weight loss for the “25%” O -ethylxanthate precursor polymer  19  was 25% with the
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks