National Sun Yat-sen University Institutional Repository:Item 987654321/38102

行政院國家科學委員會專題研究計畫 期中進度報告

新穎嵌段共聚物之自組奈米結構(2/3)

計畫類別： 整合型計畫

計畫編號： NSC 92－2216－E－110－009

執行期間： 92 年 08 月 01 日 至 93 年 07 月 31 日

執行單位： 國立中山大學材料科學研究所

計畫主持人： 蘇安仲

共同主持人： 蔡敬誠，廖建勛，陳信龍，陳文章，何榮銘

中 華 民 國 93 年 5 月 23 日

報告類型： 精簡報告

報告附件： 出席國際會議研究心得報告及發表論文

處理方式： 本計畫可公開查詢

第 1 頁，共 1 頁

2004/5/23

https://nscnt12.nsc.gov.tw/prreport_new/email/Preview.asp

新穎嵌段共聚物之自組奈米結構

NSC92-2216-E-110-009

蘇安仲 2004.03.27.

本計畫以新穎嵌段共聚物之形態分析與調控為主軸，結合學門內熟悉超分子結

構分析技術之三位學者(中山大學蘇安仲教授、清華大學陳信龍教授及何榮銘副教授)

與具有精準合成嵌段共聚物能力之三位學者(台灣大學陳文章教授、中正大學蔡敬誠

副教授、元智大學廖建勛助理教授)，針對多項子題作合作探索，以期能在尖端領域

(包括奈米材料與光電高分子)發展出具有突破能力的整合研究團隊。目前主要的成

果與進度如下述。

子題一､聚氧乙烯接枝聚苯胺之超分子自組行為與異方性導電性質 [廖建勛/陳信

龍]。 本研究將低分子量聚氧乙烯鏈端接上磷酸基(簡稱為 PEOPA)，再將其以離子

鍵連接在 PAn 主鏈上而形成一水溶且導電的超分子接枝共聚體(PAn-g-PEOPA)，發

現其具有溶致變色(lyochromic)及時致變色(chronochromic)的效應，此應與其自組裝

特性有關。我們針對其自組裝結構與 PEOPA 側鏈在奈米微相環境中的結晶行為作系

統性探討。根據小角 X 光散射之解析，PAn-g-PEOPA 自組裝成一維堆疊層狀與六方

堆積圓柱之奈米級結構，此結構在 PEOPA 結晶之後仍可被完整維持。PEOPA 側鏈

在微相環境之結晶動力學受到奈米尺度侷限的影響，隨接枝密度減少而大幅下降，

在低接枝密度時甚至出現均質孕核控制之機制。我們目前正針對自組裝結構與異方

性離子導電度之關聯性進行探討。

子題二､聚苯乙烯-聚(2-乙烯吡啶)段式共聚體模製無機奈米孔隙材料之研究 [陳文

章/陳信龍/黃慶怡]。

本研究採用活性陰離子聚合法成功製備線形與星形聚苯乙烯-聚(2-乙烯吡啶)段式共聚體(PS-b-P2VP)，並可控制段長、段比、及微結構形態。我

們利用線形 PS-b-P2VP /三氧化二矽烷(MSSQ)混成材料作為模版，製備具有長程規

則性且孔隙尺寸分佈狹窄之奈米孔隙材料。我們由小角 X 光散射(SAXS)強度之大幅

增強，直接證明本系統在高溫處理後確實產生奈米孔隙。經由對 SAXS 圖譜之結構

因子與形狀因子分析，我們定量解析各種製程條件(包括段式共聚體之分子量、PS

段之體積分率、PS-b-P2VP/MSSQ 混成比例)所製備之奈米孔隙材料奈米孔之排列規

則性、幾何形狀、平均尺寸、與奈米孔徑分佈等。此一結論並經穿透式電子顯微鏡

與原子力顯微鏡之形態觀察以及自洽均場理論計算之輔助驗證。我們目前正更進一

步對星形段式共聚體之自組裝結構及其所產生之奈米孔隙材料結構進行研究。

子題三､奈米尺度空間侷限下段式共聚體之摻合體中鏈長篩分結晶行為之抑制 [陳

信龍/蔡敬誠/廖建勛]。 分子量分佈寬廣的高分子常有鏈長篩分結晶(molecular

weight fractionated crystallization)現象，即長分子鏈會先結晶而與短分子鏈分離(亦即

分子量差異大的分子鏈無法共結晶)。我們將一對稱型與一不對稱型之聚氧乙烯-聚

丁二烯雙段共聚體 (PEO-b-PB)摻合，小角 X 光散射圖譜顯示均勻的層板微相結構。

有趣的是，長與短的 PEO 鏈段在任何結晶溫度下都會共結晶，此與其對應之聚氧乙

烯摻合體所呈現的鏈長篩分結晶行為完全不同。我們的解釋是：此一奈米尺度空間

侷限下之共結晶現象是由與 PEO 連結之 PB 鏈段所誘導的，即 PB 微相中長與短的

PB 鏈段須維持均勻混合以保持足夠的構形自由度，由於 PB 鏈段對 fractionation 的

抗拒，迫使 PEO 段必須共同結晶，因而抑制了鏈長篩分機理之展現。

子題四､硬性與軟性奈米尺度空間侷限下段式共聚體之結晶行為 [何榮銘/蔡敬誠]。

本研究整合配位聚合與陰離子聚合反應技術，經由 dimethylsilyl(cyclopentadienyl)

(9-flurenyl)zirconium dichloride/MAO 觸媒系統，在使用對甲基苯乙烯為鏈轉移劑的

條件下，成功的合成 telechelic 對排聚丙烯，並藉由陰離子偶合反應獲得不同段長比

例之對排聚丙烯-雜排聚苯乙烯(sPP-b-PS)。此一段式共聚體特殊之處在於 PS 與 sPP

鏈段具有中等程度的分離趨勢，且 PS 之玻璃轉化溫度(T

)剛好落在 sPP 之結晶溫

度範圍的正中。針對此一系統之小角 X 光散射與穿透式電子顯微鏡分析結果顯示：

結晶溫度低於 T

時，sPP 之結晶屬於 confined 形態，侷限於硬性之 sPP-b-PS 微相

結構；結晶溫度接近 T

時，sPP 之結晶轉為 templated 形態，暗示分子鏈之移動能

力開始主導結晶行為；結晶溫度超過 T

後，sPP 之結晶掙脫奈米尺度空間侷限而

成為 breakout 形態。此為文獻中首度在單一材料中觀察到三種不同的空間侷限結晶

形態，清楚的說明了侷限環境之剛性對於奈米尺度微相中結晶行為之影響。

子題五､對排聚苯乙烯之結晶行為與接枝雜排聚苯乙烯側鏈之影響 [蔡敬誠/蘇安

仲]。 本研究先以整合配位聚合技術，採用典型之 Cp*Ti(OBu)

觸媒，合成高立體

規則度的對排聚苯乙烯或其與對甲基苯乙烯之雜式共聚體；其後再以 BuLi 脫去一個

甲基氫並藉以引發苯乙烯之活性陰離子聚合反應，進而得到不同雜排聚苯乙烯側鏈

長度與接枝率之對排聚苯乙烯(sPS-g-aPS)。我們以微差掃描卡計、偏光顯微鏡、掃

描式電子顯微鏡、穿透式電子顯微鏡、寬角 X 光繞射、與小角 X 光散射等技術仔細

分析其結晶行為，得到下列幾個結論： (1) 就對排聚苯乙烯單聚體而言，屬於三角

晶系之α相實為高溫穩定晶相(其平衡熔點約為 291

_{C)，屬於正交晶系之β相則為高}

溫穩定晶相(其平衡熔點約為 284

C)。由於晶面能與結晶熱之差異，α相之熔點受晶

板厚度之影響較大；在晶板厚度小於 5 nm 時，此二相之相對穩定度反轉，造成先前

文獻中以α相為高溫相之誤解。 (2) 此二相之晶板皆具高度剛性，不易分支與扭轉，

導致球晶在數十微米大小仍未完成球狀對稱形態。先前文獻中所報告之特殊負光性

球晶，實為尚未完全發展之鞘形(sheath-like)先驅體在軸向觀察時之誤判，其構造實

無特殊之處。 (3) 對甲基苯乙烯之雜式共聚或雜排聚苯乙烯側鏈之接枝皆傾向於壓

抑β相之生成，並造成α相之熔點(與平衡熔點)之大幅下降以及球晶成長速率之明確

減緩。 (4) 在高結晶溫度範圍(Regime II)，對排聚苯乙烯之球晶成長速率之隨主鏈

長度或側鏈長度與接枝度之減緩，大體可以 de Gennes 之蛇行模型予以解釋。

子題六､共軛段式共聚體之分子疊積與發光性質 [蘇安仲/廖建勛]。 本研究結合芳

香環偶合氧化聚合反應與活性自由基聚合反應，獲得不同鏈段比例之聚苯乙烯-聚二

辛茀-聚苯乙烯(PS-PFO-PS)三段共聚體以及作比較用的聚二辛茀單聚體(PFO)。我們

以微差掃描卡計、偏光顯微鏡、穿透式電子顯微鏡、寬角 X 光繞射、吸收光譜與螢

光光譜等技術仔細分析其相行為與與發光性質之間的關係，得到下列幾個結論： (1)

PFO 基本上有高序(α)晶相、低序(α')晶相、高溫向列(N)介相、與溶致層列(β)介相等

四種相結構；經由加工條件之調整，我們可以控制此四相之生成，並定出其吸收與

發光特徵。 (2) 經由為米尺寸單晶之培養與擇區電子繞射，我們定出α相之結晶構

造(正交晶系，8 鏈晶胞，空間群 P2

2

2

，晶格常數 a = 2.56 nm, b = 2.34 nm, c = 3.32

nm)與分子疊積方式，並據以解釋α相之發光特性。(3) PS-PFO-PS 三段共聚體中 PFO

鏈段之結晶被抑制，澆注膜以β介相為主，發射波長為 440 nm；高溫熱處理後轉為

N 介相，並產生大量 520 nm 以上之 excimer emission。

Supramolecular Structure and Crystallization Behavior of a ω-Methoxy Poly(ethylene oxide) Phosphate Ester Doped Polyaniline

Bhanu Nandan1_{, Hsin-Lung Chen}1_{* and Chien-Shiun Liao}2

1 _{Department of Chemical Engineering, National Tsing Hua University, Hsin-Chu, Taiwan 30013, R.O.C.} 2 _{Department of Chemical Engineering, Yuan Ze University, Nei-Li, Taoyuan, Taiwan 320, R.O.C.}

Abstract: The supramolecular structure of a hairy-rod polymer consisting of ω-methoxy poly(ethylene oxide)

phosphate ester doped polyaniline complex (PEOPA-PANI) has been investigated using small-angle X-ray scattering (SAXS). The structure formation has been further correlated with the crystallization behavior of poly(ethylene oxide) in these complexes, studied using differential scanning calorimetry. In the case of low-molecular weight (LMW) PEOPA doped PANI complexes an imperfect ordered supralow-molecular structure was observed near the stoichiometric and lower doping ratios, whereas for higher doping ratios, macrophase separation was identified. The formation of mesomorphic structure in these complexes significantly affects the crystallization behavior of poly(ethylene oxide) side chains. The peak crystallization temperature shifts to lower temperatures, whereas the degree of crystallinity related to heat of fusion decreases appreciably with the decrease in doping molar ratios. Further the SAXS profiles show an increase in the long period with doping molar ratio since increasing side chain density in the mesophase provides less interfacial area per side chain which leads to a higher stretching of the PEO chain normal to the interface. High temperature SAXS profile shows an order-disorder transition from microphase separated morphology. The disordered morphology of LMW PEOPA-PANI complexes was found to exhibit a surface fractal behavior. Similar studies carried out on a high molecular weight (HMW) PEOPA doped PANI complexes show a significantly different behavior. In this case macrophase separation was predominant as inferred from the DSC data where the parameters are almost constant with composition and from SAXS profiles which shows a continuous scattering profile for samples in the melt state. The anomalous behavior has been attributed to the poor doping efficiency of HMW PEOPA.

Introduction

The process of self-organization facilitates formation of ordered polymeric nanoscaled structures. It is well known that block copolymers form self-organized nanoscaled structures due to repulsion between the covalently connected blocks.1-3 _{Similar to block copolymers are polymers with comb-shaped architecture which}

also has a tendency to self-organize in essentially similar way.4 A special case consists of so-called hairy rods where there are flexible side chains covalently bonded to rigid or semirigid backbones.5,6 _{Here, the comb-shaped}

architecture is of special importance as it also induces fusibility and solubility in a straightforward way to the otherwise intractable backbone. In comb-shaped polymers, the attractive covalent, i.e., permanent, interaction between the backbone and the repulsive side-chains can be replaced by weaker interactions such as ionic interactions or physical interactions such as hydrogen bonding to form supramolecules which may analogically lead to self-organized or mesomorphic nanostructures.7-17 _{Application of this concept to rod-like polymers such}

as conjugated electroactive polymers would be interesting as it could lead to possible new applications. The experimental results on the supramolecular behavior of hairy-rod polymers are scarce and only recently some theoretical aspects of there phase behavior have been reported.15,16,18,19

Recently, a water soluble conducting polyaniline has been prepared using a phosphoric acid terminated poly(ethylene oxide) (PEO) i.e. ω-methoxy poly(ethylene oxide) phosphates (PEOPA), the acid part of which act as the protonic acid dopant for PANI main chain and the long hydrophilic tail render the complex water soluble.20,21 _{Here, the rod like character of polyaniline main chains coupled with the presence of aliphatic side}

chains creates a hairy-rod type of polymer system which may have a tendency to self-organize into supramolecular structures. In the present work we have investigated the supramolecular structure formation in these water soluble hairy-rod polyanilines. The side chains in this hairy rod polymer is the oligomeric poly(ethylene oxide) which has a tendency to crystallize. The crystallization behavior of these PEO chains in the hairy rod PANI has also been studied which provides additional information about the molecular organization. Further the effect of side chain molecular weight on the structure formation in these hairy rod polymers has also been investigated. We have presented our results here in two parts; in the first part the crystallization behavior and bulk structure of a low molecular weight (LMW) PEOPA doped PANI will be presented, where we speculate the formation of microphase separated self-organized structure whereas in the second part results related to high molecular weight (HMW) PEOPA doped PANI will be presented which essentially has a macroscopic phase separated structure.

Experimental Material Preparation

ω-Methoxy poly(ethylene oxide) phosphates (PEOPA) was synthesized from poly(ethylene glycol) monomethyl ether (PEGME) (Mn = 550 (LMW) and 2000 (HMW)) and phosphorus pentaoxide using a reported

method shown in Scheme 1. As shown in the scheme the product was a mixture of mono- and bi-hydroxyl acids with a molar ratio of 1:1. So the average molecular weight of the PEOPA dopants was estimated to be 896 and 3071 gmol-1 for PEGME-550 and PEGME-2000 respectively. The aqueous solution of conducting polyaniline was prepared by adding stoichiometric ratio of PANI and the phosphate ester i.e. PEOPA into distilled water and stirring for 72 hours. The structure of PEOPA doped PANI is shown in Scheme 2. The conducting film of the PEO-PANI complex was obtained from the preceding solution by casting it in a Petri dish, drying on a hot plate at 50oC and then in a vacuum oven at 40oC for 24hrs. The stoichiometric molar ratio of PEOPA and PANI taken in the present study and their corresponding weight fractions are given in Table 1.

Characterization

Differential Scanning Calorimetry: The non-isothermal crystallization studies were carried out using a TA Instrument 2000 differential scanning calorimeter equipped with the RCS cooling system. Calibration for temperature and heat flow was made prior to sample analysis using Indium as the standard. All measurements were carried out with the sample in nitrogen atmosphere. Sample size was maintained around ~3.5mg. Each sample was first heated to 75oC, which is well above the melting point of PEO, and then was kept at this temperature for 5 minutes to ensure complete melting of the crystals. The samples were then cooled at constant cooling rates (1, 2, 5, and 8o_{C/min). The exothermic crystallization peak was then recorded as a function of}

temperature. The melting endotherms were than recorded by heating the same samples at a rate of 10o_C/min.

SAXS Measurements: SAXS characterization of all samples was conducted at 90oC to establish their melt structure. The X-ray source of SAXS, a 1.5kW X-ray generator equipped with a Cu tube, was operated at 35 mA and 40 kV. The incident X-ray beam was monochromated by a pyrolytic graphite, and a set of two pinhole inherent collimators were used so that the smearing effects inherent in slit-collimated small-angle X-ray cameras can be avoided. The scattered intensity was detected by a two-dimensional position-sensitive detector (Bruker AXS) with 512 × 512 channels. All data were corrected by the empty beam scattering and the sensitivity of each pixel of the area detector. The area scattering pattern has been radially averaged to increase the photon counting efficiency compared with one-dimensional linear detector. The intensity profile was output as the plot of the scattering intensity (I) vs the scattering vector, q = 4π/λ sin(θ/2) (θ = scattering angle).

Results and Discussion

Low molecular weight (LMW) PEOPA doped PANI

Figure 1(a) shows the crystallization exotherms for different compositions cooled at a rate of 5o_{C/min. A}

well defined crystallization exotherm can be clearly observed in all the compositions except at the molar ratio 0.2, where the crystallization exotherm was too weak to be clearly observed. From Figure 1(a), it can be clearly observed that PEO and its corresponding phosphate ester follow a different crystallization behavior under similar conditions and the difference was prominent at slower cooling rates (not shown here). Pure PEO shows a single crystallization peak characteristic of a single step process at both slower and higher cooling rates. A very fast primary crystallization process, succeeded by a comparatively slower secondary crystallization process can be clearly observed for pure PEO. In comparison, its corresponding phosphate ester i.e. PEOPA shows a crystallization exotherm with multiple peaks. At least three different processes can be clearly distinguished for the samples cooled at slower cooling rates. The multiple crystallization exotherm in PEOPA may be appreciated if we look into the structure of the PEOPA used and it can be clearly observed that it is a mixture of two different structures. One is a linear PEO with the one end terminated by a bi-hydroxyl phosphoric acid, whereas the other is a two arm PEO connected by a mono-hydroxyl phosphoric acid which in one way act as a chain defect between the PEO chains. Now in the PEOPA 550, both types of chains will crystallize in an extended chain crystal. But since one of the chains has a central defect with the two arms positioned at 120oC with respect to each other, its crystal growth rate will be much different then the single arm PEOPA. The multiple crystallization exotherm seen in case of PEOPA may be because of this factor. Also, the crystallization exotherm in PEOPA is much broader in comparison to pure PEO. The shifting of the induction temperature to a lower value signifies the slowing of the crystallization process in PEOPA as compared to PEO and is because of chain defect in the former which hampers the crystal growth and hence lowers the overall crystallization rate. In case of PEOPA-PANI complex, a single exothermic peak was observed for all compositions at all cooling rates, which shows that the crystal growth probably is taking place through a single mechanism, in contrast to that observed in pure PEOPA. The exothermic peak shifts progressively to lower temperatures with increase of PANI content. This is clearer in Figure 1(b), which shows the variation of peak crystallization temperature (Tp)

with composition at different cooling rates. The Tp of PEOPA doped PANI at R=1.0 is almost equal to that of

pure PEOPA which reveals a macrophase separated morphology for this composition. A significant decrease in the Tp can be observed as the molar ratio R decreases further from its value of 1.0. AT R=0.3, the degree of

undercooling is higher by about 19o_{C compared to neat PEOPA and by about 25}o_{C compared to neat PEO. This}

reveals that as the molar ratio of PEOPA doped PANI decreases from R=1.0, it self-organizes in a structure whose morphology is completely different from the one observed at R=1.0.

Figure 2(a) shows the melting endotherms of LMW PEOPA doped PANI obtained on sample cooled at a rate of 1o_{C/min. Neat PEO and PEOPA show multiple melting endotherms apparently because of molecular}

weight fractionation.22 _{The higher melting peak in neat PEOPA appears at higher temperature compared to that}

in neat PEO which may be because of a different lamellar thickness, though the exact reason for this is still unclear to us. In PEOPA doped PANI, compositions at all molar ratios show a single melting endotherm except at R=1.0, where another melting endotherm is present as a faint shoulder to the main peak. The peak melting temperature at R=1.0 is close to that of pure PEOPA signifying a macrophase separation at this composition. As the molar ratio decreases further the peak melting temperature shifts towards that of neat PEO. At R=0.5 and 0.3, the peak melting temperature is same as those of neat PEO, whereas at R=0.2 it further decreases. The decrease of peak melting temperature at R=0.2 may be associated with the entropic loss of PANI semirigid chain.23 _This

entropy loss is larger for smaller values of R because more PANI segments are involved per same amount of PEOPA molecules. Figure 2(b) shows the variation of heat of fusion with composition for LMW PEOPA doped PANI. The curve is almost similar to that observed for the variation of peak crystallization temperature with composition. The heat of fusion for neat PEOPA is less than that for pure PEO because of the presence of bulky end group in PEOPA which may give more defects in the crystal structure. On doping with PANI, the heat of fusion decreases slightly at R=1.0 but than shows a significant drop at R=0.7 after which it further decreases. Obviously at R=1.0 because of macrophase separation, heat of fusion is close to neat PEOPA. The slight decrease may be because of the steric hindrance from rigid PANI chains.23 _{The sudden drop in heat of fusion at}

R=0.7 can again be related to the formation of a assembled structure at this composition. The self-assembled structure restricts the PEO chain in microdomains, where the spacing between PEO chains and rigid nature of PANI chains might not allow all of the side chains to participate in crystallization. This will reflect in the heat of fusion normalized to PEO weight fraction, which will decrease with R, as observed in our case.

To obtain information about the presence of mesomorphic structure in PANI-PEOPA complexes, small angle X-ray scattering experiments were performed on the bulk samples. Figure 3 shows SAXS profiles of LMW PEOPA doped PANI at room temperature. Since at room temperature LMW PEO is in melt state, the SAXS profiles represent the melt structure of the sample. The SAXS profile for composition R=1.0 shows a monotonically decreasing scattering curve without any significant observable peak revealing a macrophase separated structure, which corroborates the inference drawn from the DSC data. At R=0.7 and lower molar ratios, a scattering peak in the SAXS profile can be clearly observed, though no higher order scattering peak can be seen. The SAXS peak may indicate the formation of lamellar morphology in the present complexes. But it has been reported that the complexes like the present one, in the disordered state may exhibit the characteristic block copolymer like concentration fluctuation (correlation hole effect), which can also lead to a SAXS peak.14

If the peak in the SAXS profile is because of concentration fluctuation, it should follow the Ornstein-Zernicke equation24 _{for concentration fluctuation in the intermediate or high q region. In that case, the slope of a}

logarithmic plot of I(q) vs. q should have a slope of -2 at high q. In the present case the slope of the logarithmic plot of I(q) vs. q at high q was found to be -4, which is normally observed for scattering from two phase region with sharp interface given by Porod’s law. Hence, a layered structure in PANI-PEOPA complexes can be assumed. It has been shown by Lipatov and co-workers25,26 _{that presence of a single scattering peak can be}

argued to be due to a layer structure characterized by an imperfect one-dimensional order normal to the layers. Their analysis in terms of the one-dimensional correlation function demonstrated that the absence of second-order maxima is due to the correlation in the packing being restricted to approximately five adjacent layers of polymers. So assuming a mesomorphic layered structure in the present case, a long period (L) can be derived using Bragg’s law. The long period calculated using Bragg’s law has been presented in Figure 4 as a function of molar ratio, R. The long period at R=0.7 was found to be 6.54 nm and it decreases with the decreasing molar ratio. The value of long period in PEOPA-PANI complexes suggests that the PEO side chains are considerably stretched and also that they are arranged in bilayers in the mesophase with very little interdigitation. Further, a decrease in the long period with decrease in R suggests that the side chains become progressively less stretched. This may be because of the fact that with decrease in R, interfacial area per PEO chain increases in the PEO microdomain, which will allow the chains to relax more freely in order to gain the conformational entropy and hence the long period will decrease. Also it must be mentioned here that at R=0.5 the doping of PANI is completed and hence at R=0.7 some free PEOPA chains will be present, which either may get hydrogen bonded to the aminic sites in doped PANI chain through hydroxyl groups or may get interdigitated with the associated i.e. doped PEOPA chains. The increase of long period in case of R=0.7 shows that some amount of free PEOPA does interdigitate between the doped PEOPA chains. Also, the SAXS scattering peak was found to become

progressively broader and of lower intensity as R decreases, which shows that at low concentrations of PEO only small parts of different chains of a somewhat higher than average degree of protonation segregate in small domains with a mesomorphic structure.8

Figure 5 shows SAXS profiles of PANI-PEOPA complex with R=0.5, at different temperatures. It can be observed from the figure that as temperature increases, scattering peak shifts toward lower q, signifying an increase in the long period which may be because of the thermal expansion effect with increasing temperature. At a much higher temperature of 175o_{C, the scattering peak was found to disappear and a monotonically}

decreasing scattering profile was observed. The scattering profile at a higher temperature suggests an order-disorder transition from a microphase separated phase to an isotropic phase. In the order-disordered phase the rod and coil macrophase separates and may form individual domains. Further analysis of SAXS profile at higher temperature was done by plotting a log I(q) vs. log q plot for scattering profiles obtained at 175o_{C and is shown}

in Figure 6. The plot shows two regions of power law behavior i.e. at intermediate and higher q. The slope of the curve at intermediate q was found to be -1 and corresponds to the rod like character of individual PANI chains presumably with a high persistence length. In the high q region, the slope of plot gives a value of -3.7, which corresponds to the fractal like behavior of the isotropic phase and gives a fractal dimension of 2.3, in agreement with surface fractal behaviour.

High molecular weight (HMW) PEOPA doped PANI

The crystallization behavior and the microstructure of HMW PEOPA doped PANI was found to be completely different from that of LMW PEOPA doped PANI. Figure 7 shows the crystallization exotherms for different compositions of HMW PEOPA doped PANI obtained at a cooling rate of 5o_{C/min. As can be observed}

from the figure, the crystallization behavior of PEO in the HMW PEOPA doped PANI does not show any systematic variation and shows a behavior almost similar to that of pure components. Apparently three points should be considered as the molecular weight of PEOPA is increased in the PEOPA-PANI complexes. First, at the same molar ratios weight fraction of PEOPA in the blend will be higher for the higher molecular weight PEOPA. As shown in Table 1, the weight percent of PEOPA in the present case is almost 90% even at a low molar ratio of 0.2. Second, as the molecular weight of PEOPA will increase, fraction of PEOPA which is taking part in the doping process will decrease because of the diffusional hindrance imposed by the larger PEO chains.27 _{And thirdly, increase in the molecular weight of PEO chains will also impose conformational}

hindrance on the formation of any specific ordered arrangement of the PANI chains.27 The combined effect of above factor should be that the crystallization of PEO here will follow a process similar to that in neat PEO or PEOPA. Figure 8(a) shows the melting behavior of the different compositions of HMW PEOPA doped PANI samples cooled at 5o_{C/min. All the compositions show a single melting transition, the peak position of which}

was found to remain almost constant near the neat PEO. Similarly the variation of heat of fusion with composition, shown in Figure 8(b) suggests that in the HMW PEOPA doped PANI; PEO chains behave more or less similar to neat PEO. This clearly suggests a macrophase separated morphology in HMW PEOPA doped PANI.

This becomes clear from the SAXS scattering profiles obtained on these samples in the melt state at 70o_{C and shown in Figure 9. All the compositions of HMW PEOPA doped PANI show a continuous SAXS}

scattering profile with no observable peak, which clearly reveals a macrophase separated structure in this class of material. Figure 10 shows the SAXS scattering profiles of HMW PEOPA doped PANI obtained at room temperature where the samples are in crystallized state. The SAXS profiles shows a primary scattering peak at q = 0.486 and sharp higher order peaks corresponding to the lamellar structure of PEO crystallites. The positions of scattering peaks are constant with composition. Further the long period in the samples calculated from the position of primary peak was found to be 12.9 nm which is close to the extended chain conformation of PEO chains with the given molecular weight (12.7 nm).

Conclusions

The crystallization and morphological behavior of a water soluble ω-methoxy poly(ethylene oxide) phosphate ester doped polyaniline have been characterized to identify the presence of supramolecular structure in these conducting polymers. The results obtained shows that the LMW PEOPA doped PANI has a tendency to self-organize in supramolecular structures at and below a molar ratio of R=0.7. The crystallization and melting parameters shows a significant shift at this composition revealing a confined geometry because of microphase separation. The SAXS profiles does show a macrophase separation at R=1.0, whereas microphase separation below this composition. The broadness of the scattering peaks shows absence of long range order which may be understood from the fact that both rod and coils in this case are polar and hence the repulsion between the two will not be very strong. A decrease in the long period with decrease in the molar ratio in LMW PEOPA doped PANI observed was apparently because the higher interfacial area available for PEO chains at lower R allow them to relax from the stretched conformation at higher molar ratios. Also at higher temperature, SAXS profiles

reveal an order-disorder transition from microphase separated morphology. In the disordered state the PANI shows a rod like character, whereas at a more local scale the PEOPA-PANI complex show surface fractal behavior with a fractal dimension of 2.3.

Studies carried out on HMW PEOPA doped PANI shows a significantly different behavior in this class of material. Here macrophase separation is predominant as inferred from the DSC data where the crystallization and melting parameters are almost constant with composition and from SAXS data which shows a continuous scattering profile for samples in the melt state.

Acknowledgment

This work was supported by the National Science Council of R.O.C. under grant NSC92-2216-E-110-009

References

1. Bates, F. S.; Fredrickson, G. H. Annu. Rev. Phys. Chem. 1990, 41, 525.

2. Hamley, I. W. The Physics of Block Copolymers; Oxford University Press: Oxford, 1998. 3. Fischer, H.; Poser, S. Acta Polym. 1996, 47, 413.

4. Plate´, N. A.; Shibaev, V. P. Comb-Shaped Polymers and Liquid Crystals; Plenum Press: New York,, 1987. 5. Ballauff, M. Angew. Chem., Int. Ed. Engl. 1989, 28, 253.

6. Lauter, U.; Meyer, W. H.; Wegner, G. Macromolecules 1997, 30, 2092. 7. Antonietti, M.; Conrad, J.; Thu¨nemann, A. Macromolecules 1994, 27, 6007.

8. Ikkala, O.; Ruokolainen, J.; ten Brinke, G.; Torkkeli, M.; Serimaa, R. Macromolecules 1995, 28, 7088. 9. Ruokolainen, J.; ten Brinke, G.; Ikkala, O.; Torkkeli, M.; Serimaa, R. Macromolecules 1996, 29, 3409. 10. Antonietti, M.; Burger, C.; Thu¨nemann, A. Trends Polym. Sci.1997, 5, 262.

11. ten Brinke, G.; Ikkala, O. Trends Polym. Sci. 1997, 5, 213.

12. Zheng, W.-Y.; Wang, R.-H.; Levon, K.; Rong, Z. Y.; Taka, T.; Pan, W. Makromol.Chem. Phys. 1995, 196, 2443.

13. Ikkala, O.; Knaapila, M.; Ruokolainen, J.; Torkkeli, M.; Serimaa, R.; Jokela, K.; Horsburgh, L.; Monkman, A. P.; ten Brinke, G. Adv. Mater. 1999, 11, 1206.

14. Chen, H. L.; Hsiao, M. S. Macromolecules 1999, 32, 2967.

15. Ruokolainen, J.; Eerikainen, H.; Torkkeli, M.; Serimaa, R.; Jussila, M.; Ikkala, O. Macromolecules 2000, 33, 9272.

16. Ruokolainen, J.; Lahtinen, M.; Torkkeli, M.; Serimaa, R.; Valkonen, J.; Rissanen, K.; Ikkala, O. Macromolecules 2001, 34, 7789.

17. Chen, H. L.; Ko, C. C.; Lin, T. L. Langmuir 2002, 18, 5619.

18. Stepanyan, R.; Subbotin, A.; Knaapila, M.; Ikkala, O.; ten Brinke, G. Macromolecules 2003, 36, 3758. 19. Knaapila, M.; Stepanyan, R.; Horsburgh, L. E.; Monkman, A. P.; Serimaa, R.; Ikkala, O.; Subbotin, A.;

Torkkeli, M.; ten Brinke, G. J. Phys. Chem. B 2003, 107, 14199.

20. Geng, Y.H.; Sun, Z.C.; Li, J.; Jing, X.B.; Wang, X.H.; Wang, F.S. Polymer 1999, 40, 5723.

21. Wang, Y.; Wang, X.; Li, J.; Zhang, H.; Mo, Z.; Jing, X.; Wang, F.; J. Polym. Sci. Polym. Phys. 2002, 40, 605.

22. Point, J.J. Macromolecules 1997, 30, 1375.

23. Luyten, M. C. Alberda van Ekenstein, G. O. R.; ten Brinke, G.; Ruokolainen, J.; Ikkala, O.; Torkkeli, M.; Serimaa, R. Macromolecules 1999, 32, 4404.

24. Ornstein, L. S.; Zernike, F. Proc. Acad. Sci. Amsterdam 1914, 17, 793.

25. Shilov, V. V.; Tsukruk, V. V.; Bliznyuk, V. N.; Lipatov, Yu. S. Polymer 1982, 23, 484. 26. Shilov, V. V.; Tsukruk, V. V.; Lipatov, Yu. S. J. Polym. Sci., Polym. Phys. Ed. 1984, 22, 41. 27. Chen, S.A.; Lee, H.T. Macromolecules 28, 2858, 1995.

Table 1. Various compositions of PEOPA/PANI used in the present study, in terms of molar ratio and weight percent ratio.

Molar Ratio (R)

R=[PEOPA]/[An] Weight Percent Ratio (PEOPA/PANI),

When Mol. Wt. PEOPA = 896g/mol-1

Weight Percent Ratio (PEOPA/PANI),

When Mol. Wt. PEOPA = 3071g/mol-1 R =1.0 90.8/9.2 97.1/2.9 R=0.7 87.3/12.7 95.9/4.1 R=0.5 83.1/16.9 94.4/5.6 R=0.3 74.7/25.3 91.0/9.0 R=0.2 66.3/33.7 87.1/12.9 P2O5 + 3CH3(OCH2CH2)nOH

CH3(OCH2CH2)nOPO(OH)2 + [CH3(OCH2CH2)nO]2PO(OH)

P

O

O

OH

OH

P

O

O

H

O

O

-50 -40 -30 -20 -10 0 10 20 30 40 Temperature (0 C) Endo Exo PEO PEOPA R=1 R=0.7 R=0.5 R=0.3 R=0.2 PEO 100 90 80 70 60 -30 -25 -20 -15 -10 -5 0 5 10 P e ak C ry s ta ll iz at io n T e m p er a tu re (Tp ), oC

Weight Percent PEOPA

1oC/min 2oC/min 5oC/min 8o C/min PEO 100 90 80 70 60 -30 -25 -20 -15 -10 -5 0 5 10 P e ak C ry s ta ll iz at io n T e m p er a tu re (Tp ), oC

Weight Percent PEOPA

1oC/min 2oC/min 5oC/min 8o C/min

(

(

Figure 1. (a) Crystallization exotherms at a cooling rate of 5oC/min for LMW PEOPA doped PANI; (b)

Variation of peak crystallization temperature as a function of weight percent PEOPA for LMW PEOPA doped PANI at different cooling rates.

80 on ( J /g 80 on ( J /g -40 -30 -20 -10 0 10 20 30 40 50 Temperature (o C) Endo Exo PEO PEOPA R=1 R=0.7 R=0.5 R=0.3R=0.2 PEO 100 90 80 70 60 0 20 40 60 100 120 140 H e at o f F u s i )

Weight Percent PEOPA

PEO 100 90 80 70 60 0 20 40 60 100 120 140 H e at o f F u s i )

Weight Percent PEOPA

(

(

Figure 2. (a) Representative melting endotherms at a heating rate of 10oC/min for LMW PEOPA doped PANI

after being cooled at a rate of 1o_{C/min; (b) Variation of heat of fusion as a function of weight percent PEOPA}

0.5 1.0 1.5 2.0 2.5 3.0 Inte nsi ty ( a .u .) q (nm-1 ) R=1 R=0.7 R=0.5 R=0.3 R=0.2

Figure 3. SAXS profiles of LMW PEOPA doped PANI at room temperature (melt structure).

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 Long Pe ri od, L (nm ) R

Figure 4. Variation of long period with composition for LMW PEOPA doped PANI complexes.

0.5 1.0 1.5 2.0 2.5 3.0 In te n s it y (a .u .) q (nm-1) 1750C 70oC RT

1 1 10 100 lo g (I (q )) log (q) -1 -3.7

Figure 6. SAXS intensity plotted as a function of q on a logarithmic scale for R=0.5 at 175oC.

10 20 30 40 50 Temperature (o_C) PEO PEOPA R=1 R=0.7 R=0.5 R=0.3 R=0.2 Endo Exo

Figure 7. DSC crystallization exotherms at a cooling rate of 5oC/min for HMW PEOPA doped PANI complexes.

30 40 50 60 70 Temperature (o C) Endo PEO PEOPA R=1 R=0.7 R=0.5 R=0.3 R=0.2 PEO 105 100 95 90 85 130 140 150 160 170 180 190 H eat o f Fus ion ( J /g)

Weight Percent PEOPA

Figure 8. (a) Representative melting endotherms at a heating rate of 10oC/min for HMW PEOPA doped PANI

after being cooled at a rate of 1o_{C/min; (b) Variation of heat of fusion as a function of weight percent PEOPA}

0.5 1.0 1.5 2.0 2.5 In te n s ity (a .u .) q (nm-1) R=1 R=0.7 R=0.5 R=0.3 R=0.2

Figure 9. SAXS profiles of HMW PEOPA doped PANI at 70oC (melt structure)

0.3 0.6 0.9 1.2 1.5 1.8 2.1 In te n s it y ( a .u .) q (nm-1) R=1 R=0.7 R=0.5 R=0.3 R=0.2

Cocrystallization Behavior in Binary Blend of Crystalline-Amorphous Diblock Copolymers

Yen-Yu Huang1_{, Bhanu Nandan}1_{, Hsin-Lung Chen}1_{*, Chien-Shiun Liao}2 _{and U-Ser Jeng}3

1_{Department of Chemical Engineering, National Tsing Hua University, Hsin-Chu, Taiwan 30013, R.O.C.} 2_{Department of Chemical Engineering, Yuan Ze University, Nei-Li, Taoyuan, Taiwan 320, R.O.C.} 3_{National Synchrotron Radiation Research Center, Hsin-Chu, Taiwan 300, R.O.C.}

*To whom correspondence should be addressed

Abstract

The binary blends of a short symmetric poly(ethylene oxide)-block-polybutadiene (PEO-b-PB) and a long asymmetric PEO-b-PB have been studied to examine the cocrystallization behavior of the longer and the shorter PEO blocks confined in the lamellar microdomains. Cocrystallization was found to occur over a wide range of undercooling whereas the corresponding blends of PEO homopolymers with similar molecular weights and compositions displayed phase-segregated crystallization. In contrast to the kinetically trapped solid solutions formed in homopolymer blends, the cocrystallization behavior observed here may be driven thermodynamically to attain a lower interfacial energy in the system while allowing the long PB blocks in the asymmetric diblock to relax conformationally.

Text

Fractionation during crystallization of heterodisperse homopolymers is a well-known phenomenon that originates from the mixing gap in the crystalline state of chain molecules differing sufficiently in length. Under thermodynamic equilibrium conditions, a binary mixture of homologous chain molecules with large disparity in molecular weights solidifies to form two solid phases. Under nonequilibrium conditions, however, solid solutions may form at relatively low solidification temperatures where the crystallization proceeds so rapidly that the cross-diffusion of the molecules, which is an essential step in the fractionation, is impeded.1

Recently, much attention has been paid to crystalline-amorphous diblock copolymers (C-b-A) and their blends with homopolymer A because their self-assembled structures provide interesting templates to explore the crystallization behavior of polymer chains confined in nanoscale space once C block forms microdomains in the melt state.2-5 _{In this study, we will deal with the blends of two chemically identical C-b-A with different chain}

lengths, i.e., Cα-b-Aβ/Cγ-b-Aδ with the subscript representing the degree of polymerization. Though the binary

blends of diblock copolymers have received considerable theoretical and experimental interests due to richness of their phase behavior, most of the works have been carried out on systems with all amorphous blocks.6-12

Introduction of a crystallizable block into this type of system is expected to create more complex phase behavior than normally has been observed for their amorphous counterparts. A very important question which arises naturally is whether Cα and Cγ blocks originally mixed within the microdomains in the melt would cocrystallize

or phase segregate into their own crystalline lamellar structures. In this study, we intend to demonstrate that cocrystallization leading to a solid solution occurs over a broad range of undercooling in the lamella-forming blends of a nearly symmetric C-b-A with an asymmetric counterpart, while the corresponding homopolymer blends with similar molecular weight and compositions exhibit phase-segregated crystallization. In contrast to the kinetically trapped solid solution found in homopolymer systems, the solid solution generated in the diblock blends may be equilibrium in nature governed by the interfacial energy and the conformational entropy of the A blocks.

The system under study is the blends of a short nearly symmetric poly(ethylene oxide)-block-polybutadiene (PEO-b-PB) and a long asymmetric PEO-b-PB with PEO and PB being the crystalline and amorphous blocks, respectively. Both copolymer samples were synthesized by the anionic polymerizations of ethylene oxide and butadiene. The nearly symmetric PEO-b-PB (denoted as E170B102) had the number average molecular weight of

the respective blocks as Mb,PEO = 7500, Mb,PB = 5500 and the polydispersity index (Mw/Mn) = 1.04. Mb,PEO and

Mb,PB of the asymmetric PEO-b-PB (denoted as E80B481) were 3500 and 26000, respectively. The melt

morphology of neat E170B102 and E80B481 is lamella and BCC-packed PEO spheres, respectively, judging from

the corresponding overall volume fractions of PEO block (i.e., 0.54 and 0.10). E80B481/E170B102 blends were

prepared by solution mixing using toluene as the solvent, followed by removing the solvent in vacuo at 80 o_C.

The compositions of the blends prepared are designated as the weight ratio of E80B481 to E170B102. The blends of

two PEO homopolymers with approximately the same molecular weights as those of the PEO blocks in the copolymer blends were also prepared for comparing the crystallization behavior of PEO in these two types of systems. The PEO homopolymers with Mh,PEO = 3350 (denoted as h-E76) and Mh,PEO = 8000 (denoted as h-E182)

were purchased from Sigma. The h-E76/h-E182 blend compositions were based on the calculated weight ratios of

the two PEO blocks in the PEO domain of the copolymer mixtures (cf. Table 1).

First we would like to verify that the shorter and the longer PEO/PB blocks in E170B102 and E80B481 mix

intimately in their respective microdomains in the melt state. Figure 1(a) shows a series of small angle X-ray scattering (SAXS) profiles obtained in the melt state from E80B481/E170B102 blends with the weight percent of

E80B481 up to 55%. Each profile is characterized by a primary scattering peak along with the higher-order

maxima situating at integer multiples of the first-order peak position, which clearly indicates the formation of lamellar morphology in these blends. Figure 1(b) plots the interlamellar distance (D) and the area per junction point at the lamellar interface (Σ) as a function of the number fraction of the asymmetric E80B481 (nas). The

interfacial area per junction was obtained from12

(1)

}/D

]v

N

n

)N

n

[(1

]v

N

n

)N

n

2{[(1

Σ

=

−

+

+

−

+

where Ns,i and Nas,i (with i = PB or PEO) is the degree of polymerization of i block in E170B102 and E80B481,

respectively, and vi is the volume of an i monomer unit (vPB = 0.095 nm3 and vPEO = 0.072 nm3). It can be seen

that the interlamellar distance increases with increasing fraction of E80B481, whereas Σ remains constant and

equals approximately to that in neat E170B102. The observed composition dependences of D and Σ are in parallel

with those exhibited by the blends of a short symmetric polystyrene-block-polyisoprene (PS-b-PI) and a long asymmetric PS-b-PI reported by Court and Hashimoto.11,12 _{The swelling of interlamellar distance upon adding}

E80B481 indicates that the bipopulated PEO/PB blocks mix intimately in their respective domains, where the

much longer PB blocks from E80B481 effectively swells the thickness of PB lamella. The interfacial area per

junction point is however dominated by the short symmetric diblock as manifested by the proximity of Σ to that in neat E170B102 irrespective of the blend composition. In this case, each lamellar domain (PB in particular) is

constituted of two layers of brushes lying on top of each other. The first layer is formed by the shorter blocks and the first subchains in the longer blocks, while the second layer composes of the remaining subchains of the longer blocks, as illustrated in Figure 2(a).12

The intimate mixing of the block chains in the respective domains is further supported by the suppression of the third-order diffraction peak for 40/60, 50/50 and 55/45 blends (cf. Figure 1(a)), in which the overall volume fraction of PEO (fPEO) lies in the vicinity of 0.3 (cf. Table 1). As the intensity of the nth-order

diffraction from the lamellar array is proportional to sin2_(nπf

PEOL) with fPEOL being the volume fraction of PEO

lamella in the array, the diminishment of the third-order peak means that fPEOL is close to 1/3. The proximity

between fPEO prescribed by the feed ratio and fPEOL attests that the lamellar stacks are formed by the uniform

mixing of the two diblocks. The intimate mixing is favored because the resultant lamellar structure has a lower interfacial energy (compared with the macrophase-separated state containing the spherical domains formed by E80B481), while allowing the long PB blocks in E80B481 to relax conformationally in the lamellar domains

according to the chain packing mode depicted in Figure 2(a).

The cocrystallization or phase-segregated crystallization phenomenon in the diblock and homopolymer blends are revealed from the melting behavior of the samples isothermally crystallized at different temperatures (Tc). The representative DSC melting curves of h-E76/h-E182 and E80B481/E170B102 blends subjected to isothermal

crystallization at 8 o_{C (high undercooling) and 40} o_{C (low undercooling) are displayed in Figure 3. From Figure}

3(a) and (b), it can be observed that most homopolymer mixtures show two melting endotherms irrespective of Tc, one of which is present as a shoulder to the main peak. The two melting peaks correspond to those of the

pure homopolymers and hence reveal a fractionation process during crystallization in the binary mixtures. The 5/95 blend tends to display a single endotherm upon crystallization at 8 o_{C, which may be due to the formation}

of a kinetically trapped solid solution in this dilute mixture at high undercooling. The molecular weight fractionation during crystallization in the binary mixtures of PEO homopolymers has been well described in the past.13-15 _{In a study of the crystallization behavior of a mixture of PEO in a narrow molecular weight range,}

Prudhumme13 _{found that the mixtures crystallized by a fast cooling showed a single melting endotherm}

indicating the occurrence of cocrystallization, whereas slow crystallization allowed fractionation of the sample. Cheng and Wunderlich14 _{investigated the phase-segregated crystallization in binary mixtures of PEO with a}

wider range of molecular weight and found that the two components tended to cocrystallize and phase segregate at high and low undercoolings, respectively. A more comprehensive study by Balijepalli et al.15 _{has drawn}

similar inferences. Though the phase-segregated crystallization behavior at low undercooling such as at 40 oC (cf. Figure 3(b)) is quite typical from the above literature reports but that observed at relatively high undercooling (cf. Figure 3(a)), is rather unexpected. This shows that the blends of relatively low molecular weight PEO have a strong tendency towards fractionation during crystallization, which prevails even at the crystallization conditions where non-equilibrium solid solution may otherwise be expected.

The melting curves of E80B481/E170B102 blends are displayed in Figure 3 (c) and (d). It is noted that the

melting curve of neat sphere-forming E80B481 shown for comparison was obtained after cooling the sample to –

50 o_{C because crystallization of PEO blocks from the spherical microdomains must predominantly involve}

homogeneous nucleation which takes place at very large undercooling (∆T > 90 K).2,3 _{Contrary to the observed}

behavior for homopolymer blends, the diblock copolymer mixtures exhibit a single melting endotherm for crystallization at both low and high undercoolings and the peak temperature decreases as the concentration of E80B481 increases. This implies that the shorter and the longer PEO blocks in E80B481 and E170B102, respectively,

cocrystallize in the lamellar microdomains irrespective of the composition and the degree of undercooling. Nevertheless, considering that the longer PEO blocks may tend to crystallize first, this crystallization could induce a macrophase separation through which the asymmetric E80B481 is segregated to form PEO spherical

microdomains before the shorter PEO blocks have a chance to crystallize. Once the spherical domains are formed, the shorter PEO blocks confined within these domains are no longer crystallizable at the prescribed crystallization temperatures; in this case, the isothermally crystallized diblock blends will exhibit a single melting endotherm associated solely with E170B102. This scenario can however be ruled out from the fact that

the melting temperatures of the blends show a marked variation with composition, revealing the crystalline lamellae constituting of both PEO blocks. To further assert the conclusion of cocrystallization, the diblock blends isothermally crystallized at 8 and 40 o_{C were cooled to –50} o_{C followed by annealing at this temperature}

for 48 hrs. In the event of the crystallization-induced macrophase separation, the low-temperature treatment would be sufficient to crystallize the shorter PEO blocks in the spherical microdomains, such that the subsequent heating curves should show two melting endotherms. The heating curves recorded after such a treatment however produces only a single endotherm for all compositions and are virtually identical with those shown in Figure 3(c) and (d). This emphatically proves that the observed single endotherm does signal cocrystallization of the shorter and the longer PEO blocks in the copolymer blends.

Now let us consider why the diblock blends tend to cocrystallize while fractionation is favored during crystallization in the corresponding homopolymer blends. Intuitively, phase-segregated crystallization will not be allowed if microdomain A in the copolymer blend is vitrified at Tc (i.e., Tg of A block > Tc), because the

immobility of junction points at domain interface prohibits any cross-diffusion involved in the fractionation of C blocks. This condition however does not apply here since the Tc’s under study are well above the Tg of PB.

Considering that PEO and PB blocks are covalently connected, fractionation of PEO blocks will invariably demix the short and long diblocks in the blends. The interfacial energy of the system will increase if the asymmetric E80B481 demixing from the crystallizing symmetric E170B102 self-assembles into spherical micelles.

In this case, phase-segregated crystallization will not be thermodynamically favored once the reduction of the free energy of crystalline PEO phase from fractionation cannot compensate the gain in interfacial energy. The increase of interfacial energy may in principle be circumvented if the fractionation process is confined within the preexisting lamellar microdomains, where the crystallites formed by the longer and shorter PEO blocks coexist within the lamellar domains upon fractionation, as illustrated in Figure 2(b). However, this lamellar structure contains an entropic penalty associated with stretching of the long PB blocks to maintain the normal density in the PB domain. In light of the energetic and entropic penalties introduced to the system by phase-segregated crystallization, the cocystallized structure shown in Figure 2(c),16 _{in which both the PEO blocks}

cocrystallize uniformly in the PEO domain with similar lamellar thickness, may become the thermodynamically stable morphology of the system. This structure allows the lower interfacial energy and the higher conformational entropy of the long PB blocks in the melt state to be largely retained with a (presumably small) energetic penalty from cocrystallization.

In summary, we have shown that the PEO blocks of different lengths in the binary PEO-b-PB blends cocrystallized over a wide range of undercooling whereas the corresponding blends of PEO homopolymers tended to show phase-segregated crystallization. In contrast to the kinetically trapped solid solutions formed in homopolymer systems, cocrystallization in the present blends may be an equilibrium requirement to attain a lower interfacial energy and a higher conformational entropy of the long PB blocks in the system. The cocrystallization observed here represents a scenario where a kinetically driven process found in homopolymer crystallization turns into a thermodynamically controlled process in the C-b-A systems. Another example is the crystalline chain folding; chain folding during crystallization in homopolymer is a kinetically driven phenomenon, whereas in the diblock copolymers introduction of an equilibrium degree of chain folding into the crystalline domain is necessary to enlarge the cross section per junction point to relax the A block chains.17,18

This work was supported by the National Science Council of R.O.C. under grant NSC92-2216-E-110-009 and also in part by the National Synchrotron Radiation Research Center under project ID 2003-3-037-1.

References and Notes

1. Smith, P.; Manley, R. Macromolecules 1979, 12, 483.

2. Chen, H.-L.; Hsiao, S.-C.; Lin, T.-L.; Yamauchi, K.; Hasegawa, H.; Hashimoto, T. Macromolecules 2001, 34, 671.

3. Chen, H.-L.; Wu, J.-C.; Lin, T.-L.; Lin, J. S. Macromolecules 2001, 34, 6936. 4. Loo, Y.-L.; Register, R. A.; Ryan, A. J.; Dee, G. T. Macromolecules 2001, 34, 8968.

5. Xu, J.-T.; Fairclough, J. P. A.; Mai, S.-M.; Ryan, A. J.; Chainbundit, C. Macromolecules 2002, 35, 6937. 6. Shi, A.-C.; Noolandi, J. Macromolecules 1994, 27, 2936.

7. Hashimoto, T.; Koizumi, S.; Hasegawa, H. Macromolecules 1994, 27, 1562. 8. Shi, A.-C.; Noolandi, J. Macromolecules 1995, 28, 3103.

9. Zhao, J.; Majumdar, B.; Schulz, M. F.; Bates, F. S.; Almdal, K.; Mortensen, K.; Hajduk, D. A.; Gruner, S. M. Macromolecules 1996, 29, 1204.

10. Sakurai, S.; Irie, H.; Umeda, H.; Nomura, S.; Lee, H. H.; Kim, J. K. Macromolecules 1998, 31, 336. 11. Court, F.; Hashimoto, T. Macromolecules 2001, 34, 2536.

12. Court, F.; Hashimoto, T. Macromolecules 2002, 35, 2566.

13. Prud’homme, R. E. J. Polym. Sci., Part B: Polym Phys. 1982, 20, 307.

14. (a) Cheng, S. Z. D.; Wunderlich, B. J. Polym. Sci., Part B: Polym. Phys. 1986, 24, 577. (b) Cheng, S. Z. D.; Wunderlich, B. J. Polym. Sci., Part B: Polym. Phys. 1986, 24, 595. (c) Cheng, S. Z. D.; Wunderlich, B. J. Polym. Sci., Part B: Polym. Phys. 1988, 26, 1947.

15. Balijepalli, S.; Schultz, J. M.; Lin, J. S. Macromolecules 1996, 29, 6601.

16. The model structures are drawn based on the SAXS results of the isothermally crystallized diblock blends from which the PEO lamellar thickness was calculated to be ca. 10.8 nm. Assuming that crystalline stems orient perpendicular to the interface,17 this gives a once folded and triply folded chain structure for the shorter (extended chain length = 22.15 nm) and the longer (extended chain length = 47.47nm) PEO blocks in E80B481 and E170B102, respectively.

17. DiMarzio, E. A.; Guttman, C. M.; Hoffman, J. D. Macromolecules 1980, 13, 1194. 18. Whitmore, M. D.; Noolandi, J. Macromolecules 1988, 21, 1482.

Table 1 Characteristics of the Binary Diblock Copolymer Blends Studied in this Work

E80B481/E170B102

(in weight percent) Weight ratio of the shorter (Mlonger (Mb,7500 = 7500) PEO blocks in the PEO b,PEO = 3500) and the

domaina

fPEOb Melt structure of

PEO domain 0/100 0/100 0.54 Lamellae 20/80 1/19 0.45 Lamellae 40/60 1/7 0.36 Lamellae 50/50 1/5 0.32 Lamellae 55/45 1/4 0.29 Lamellae 100/0 100/0 0.10 Sphere

a _{The compositions of homopolymer mixtures (h-E}

76/h-E182) were based on this ratio. b _{Overall volume fraction of the PEO block.}

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 10-3 10-2 10-1 100 101 102 103 104 105 106 107 108 109 1010 1011 0/100 20/80 40/60 50/50 55/45 1 2 3 ₄ 5 1 2 3 4 ₅ 6 1 2 3 4 3 4 5 2 3 4 5 1 2 1

R

ela

ti

v

e In

te

n

isty

(a

.u

.)

q (nm

)

E

B

/E

B

_(a)

(b)

Figure 1. (a) SAXS profiles of lamella-forming E80B481/E170B102 blends in the melt state. The SAXS

experiments were conducted at 90 o_{C; (b) variations of the interlamellar distance (D) and the area per junction}

point at the lamellar interface (Σ) with the number fraction of the asymmetric E80B481 (nas).

Crystallization

(a)

(c)

(b)

Figure 2. Schematic illustrations of the structures of E80B481/E170B102 blend: (a) the melt structure formed by the

intimate mixing of the two diblocks, where each lamellar domain is constituted of two layers of brushes lying on top of each other; (b) a crystalline structure generated by the phase-segregated crystallization, where the crystallites formed by the longer and shorter PEO blocks coexist within the lamellar domains upon fractionation. The long PB blocks are highly stretched to maintain the normal density in the PB domain; (c) the crystalline structure generated by cocrystallization. This structure allows the lower interfacial energy and higher conformational entropy of the long PB blocks in the melt state to be largely retained.

Figure 3 10 20 30 40 50 60 70 80 90 -7 -6 -5 -4 -3 -2 -1 100/0 55/45 50/50 40/60 20/80 0/100 End o th erm ic Temperature (oC) 40 50 60 70 80 1/4 (55/45) 1/5 (50/50) 1/7 (40/60) 1/19 (20/80) pure 3350 (100/0) pure 8000 (0/100) E ndothe rm ic Temperature (oC) 40 50 60 70 80 100/0 55/45 50/50 40/60 20/80 0/100 End o th erm ic Temperature (oC) 10 20 30 40 50 60 70 80 pure 3350 (100/0) 1/4 (55/45) 1/5 (50/50) 1/7 (40/60) 1/19 (20/80) pure 8000 (0/100) En d o th er mi c Temperature (oC) h-E76/h-E182 h-E76/h-E182 E80B481/E170B102 E80B481/E170B10

(d)

(c)

(b)

(a)

Figure 3. DSC melting curves of h-E76/h-E182 and E80B481/E170B102 blends obtained after isothermal

crystallization; (a) h-E76/h-E182, Tc = 8 oC; (b) h-E76/h-E182, Tc = 40 oC; (c) E80B481/E170B102, Tc = 8 oC; (d)

E80B481/E170B102, Tc = 40 oC. Each bracket in (a) and (b) indicates the corresponding composition of the diblock

blend with the same mixing ratio of the shorter and the longer PEO chains as that in the homopolymer blend. The melting curves of neat E80B481 in (c) and (d) were obtained after cooling the sample to –50 oC.

結晶性團聯共聚合物從硬相侷限到軟相侷限之結晶行為

Various crystalline textures have been identified in a unique crystallizable block copolymer system, poly(styrene) -b-syndiotactic poly(propylene) (PS-sPP), having an intermediate segregation strength and glass transition

temperature of PS (Tg,PS) located in the midst of the sPP

crystallization window. Confined morphology for crystallization of sPP was observed while the crystallization temperature of sPP

(Tc,sPP) was less than Tg,PS. Crystallization became templated

once Tc,sPP approached Tg,PS; indicating that the final

morphology was governed by the mobility of copolymer chains

instead of the segregation strength. A further increase in Tc,sPP

could lead to a breakout in nanostructure. This study revealed the crystallization behavior of block copolymers with intermediate segregation strength from hard to soft confinement for the first time.

Keywords: hard confinement, soft confinement, segregation strength.

中文摘要

本實驗主要運用鏈轉移反應（chain transfer reaction），藉由苯

乙烯類單體作為鏈移劑配合氫氣應用於sPP 的 Metalloncene 聚合反應，製備出聚苯乙烯與對排聚丙烯雙團聯共聚合物 poly(styrene)-b-syndiotactic poly(propylene) (PS-sPP)。此一特 殊 結 晶 性 團 聯 共 聚 合 物 為 中 度 分 離 強 度(intermediate segregation strength) 的系統並且其 sPP 結晶溫度範圍跨越非 結晶團聯PS 的玻璃轉化溫度。結果發現當 PS-sPP 系統從硬 相侷限到軟相侷限環境中進行結晶時其最終形態為一個有趣 的形態發展。當結晶溫度sPP (Tc,sPP)< Tg,PS時其形態被限制 (confined)。然而隨結晶溫度上升到 Tg,PS時，其形態為模板 (templated) 形態；這反映出系統的最終形態與團聯共聚合物 中分子鏈的移動性有關。隨結晶溫度Tc,sPP增加將導致奈米微 結構的破壞 (breakout)。此系統為第一個結晶性團聯共聚合 物於中度分離強度情況下探討從硬相侷限到軟相侷限環境之 結晶行為。 關鍵詞：硬相侷限，軟相侷限，分離強度 簡介 高 分 子 團 聯 共 聚 合 物(block copolymer) ， 是 由 不 互 容 (immiscible)之團聯鏈段，彼此利用化學鍵鍵結，於平衡狀 態，自組裝(self assembly)以及自有序(self ordering)形成微相 分離之形態，隨著團聯鏈段組成體積變化，可得到不同幾何 形狀的有序奈米微結構。若將其中一團聯鏈段改變為可結晶 的高分子或將兩結晶的高分子共聚合，結晶性高分子團聯的 結 晶 行 為 將 使 得 此 類 共 聚 合 物 形 成 所 謂 structure-within-structure 的特殊有序微結構，將大幅促進此 類奈米材料設計於實際應用上的多樣化。由於結晶的效果可 有效提高材質的使用溫度及機械性質，進而增加材料之穩定 性，探討可結晶性團聯鏈段之結晶行為與奈米微觀結構之形 態變化兩者間的關聯與互動性成為一熱門的題材。目前文獻 報導的主要探討重點，著眼於可結晶鏈段的結晶行為對微觀 相分離所形成之微結構的影響，一般推論微結構的變化將與 結晶所進行的溫度(Tc)、其不結晶鏈段之玻璃轉化溫度(Tga) 及 團 聯 共 聚 合 物 之 有 序 無 序 相 轉 換 溫 度(order-disorder

transition temperature, TODT)的相對應高低有關[1-8]。當結晶溫

度大於有序無序轉換溫度（Tcc>TODT>Tga）時，由於結晶的先

行進行使得後續的相分離受到限制，一般而言材質並未出現 有序的微結構[1,2]。而當系統為硬相侷限(hard confinement,

TODT>Tga>Tcc）時，結晶鏈段被限制在相分離結構間無法大規

模的移動，微結構並不會因為結晶而改變[3-5]。相對於硬相

侷限，若為軟相侷限（soft confinement, TODT> Tcc ≥Tga）系統

其最終形態的發展是比較複雜，爭議的癥結在於結晶性鏈段 的分子鏈因結晶的自我有序現象是否具有足夠大的驅動力干 擾微觀相分離的形成以及對應相連接鏈段的牽制效應的影響 程度[6-8]。最近根據 Register 等學者提出以結晶溫度（Tc） 對應的微觀相分離強度(χN)C與有序無序轉換溫度（TODT）對 應的微觀相分離強度(χN)ODT之比值(χN)C/(χN)ODT>> 3 為極強

分離強度（extremely strong segregation）時，其系統之結晶 鏈段被限制而保持原本有序形態[8]。但對於在中度分離強度 (intermediate segregation strength)系統，尤其是層板結構，目 前關於結晶後形態依然有所爭議。此微觀相分離結構因受到 結晶影響下可能會維持原本結構，或是模板(templated) 形 態、甚至被破壞。於是本實驗設計一特殊團聯共聚合物PS-sPP 的系統，其處於中度分離強度且結晶性鏈段sPP 的結晶溫度 跨越非結晶鏈段 PS 的玻璃轉換溫度，其特色則是在同一個 材料中，可以完整的探討從硬相侷限到軟相侷限以及在 PS-Tg 時，結晶行為的變化和最終形態的不同，以其探討結 晶性團聯共聚合物中化學鍵結與非結晶鏈段的鏈移動性跟結 晶後最終形態之相關性。 實驗 一､實驗材料 由苯乙烯類單體作為鏈移劑配合氫氣應用於sPP的 Metalloncene聚合反應，製備出聚苯乙烯與對排聚丙烯雙團聯 共聚合物poly(styrene)-b-syndiotactic poly(propylene) (PS-sPP) 其數量平均分子量為14000（體積分率fPS=0.47），分子量分佈 為1.29，化學結構如圖一所示。 二､實驗儀器

(1)穿透式電子顯微鏡（Transmission Electron Microscope， TEM；型號 JEOL JEM-1200CXII）

將材料經由不同條件的處理後，利用超薄切片機切出厚

度約50-70nm 的薄片，使用染色劑以 RuO4染色PS 鏈段(黑

色部分：PS，白色部分：sPP)，利用 TEM 進行形態觀察。 (2)小角度 X 光散射儀（Small-angle X-ray scattering, SAXS）

本實驗所進行的同步輻射之SAXS 的實驗，是藉由美國

同步輻射中心(NSLS, National Synchrotron Light Source)的 X27C beamline 作為分析儀器，其中入射 X 光的波長 (λ,wavelength)為 0.1366 nm，以二維小角度 X 光散射圖譜 (SAXS pattern)中利用 silver behenate 的 first -order scattering

vector q*_=1.076nm-1_{做校正，偵測小角度X 光的儀器為}

MarCCD 偵測器。

(3)微差式掃瞄熱卡計（Differential Scanning Calorimeter， DSC；型號 Perkin-Elmer DSC 7）

在形成微觀相分離的有序微結構下，利用DSC 進行結晶

性團聯共聚合物PS-sPP 的等溫結晶與熔融行為探討。另一方