探討製程變異於有機電晶體之影響

Abstract

本論文研製之 P3HT有機電晶體，是利用溶劑式製程方式，故須將其粉末(溶質)，經選用適當的溶劑與調配比例，使得P3HT能均勻得溶於溶劑之中，再使用旋鍍製程方式，沉積形成P3HT薄膜 ; 此薄膜當成有機主動層(p-type)，進而應用於有機電子相關產品。 P3HT於有機材料中，被分類為共軛高分子；因具有自我組織的性質，而形成高分子長鏈，因製程方式不同，所得實驗結果和薄片(lamella)排列也不盡相同。經本論文實驗結果：首先藉由氯仿溶液與調配濃度百分比為0.1%之比例，能讓P3HT完全溶解於其中，不會因溶解度的關係而降低其效能；相反的，利用二甲苯卻得到相對不理想之結果，因較差的溶解度關係所造成。製程時，猜測搭配較少的比例與揮發慢的溶劑，可以提供相對較長排列時間，使得微結構分子，有較恰當的排列方式，進而得到較高之結晶性，而使的移動率相對的提高。接著探討不同規則區域性(Regioregularity_RR)和分子量(Molecular weight)，兩種溶質同樣溶於氯仿中，經電性與物性分析結果，知道分子量差異比RR值來的重要。再經由1500轉速、烘烤(Curing)溫度為150℃與時間維持3分鐘下的製程條件，具有比較好的電性結果，這也利用歐傑電子(Auger)和表面元素(ESCA)分析發現，氧分子含量都有相對比較低的情況，但是透過FTIR，並無見到氧分子所造成的結果，猜測：其的存在是利用物理吸附非鍵結，故無法被偵測到。 已知基本製程條件後，由於蕭特基接面會有較高的能位障產生，此位能障位於金屬和有機物間，然而，藉由金(GOLD)的金屬材料和適當厚度的附著層(Adhesion layer)，可因功函數接近HOMO和其化學屬性的關係，得到相對比較小阻值結果，另一方面，使用表面處理(HMDS)，讓分子可更有效排列與增加結晶性，而提高移動率與越接近零之起始電壓(Threshold Voltage)。 論文的最後，研究P3HT有機材料的穩定性，因為大氣環境對有機物來說是一種挑戰。實驗結果顯示，製程而成的電晶體，卻在存放一個月左右，即可明顯見到退化現象。可是再利用後段長壓退火，以通入氮氣方式，進行180℃和 30分鐘的退火，可以有效改善效能，這樣的效果一樣可在低分子量見到；同時發現，經過一連續製程步驟之元件，再經退火處理後，On/off 比例值可達到6~8x106次方，經由歐傑電子之縱深分析，適當退火可更有效使氧分子被驅離出薄膜;可惜的是，移動率並沒有因此而提升和起始電壓有往負值方向移動。最後利用定電壓和時間STRESS方式，發現只供給Drain -60伏電壓下，P3HT的ID電流，會隨時間上升而增加，且已達開啟電流，此時，卻未發現閘極電流隨著上升的情形。但再次量測基本電性時，發現整個OFF電流，受STRESS後，有上升趨勢，間接猜測是，因為氧分子，經由STRESS時，再次進入薄膜，而可能導致的結果。

A procedure is studying and using the solution-process to fabricate P3HT-based organic thin film transistor. Using appropriated solvent and weight percentage is needed. And then the P3HT solution will be spun and deposited to be as active layer for organic electronic. P3HT organic material is to be classified to conjugated polymer. One of advantages is a property of self-organization and based on this structure to form a long chain. This long chain will be formed a lamella structure. If this lamella were packing closely, it would result a high regioregularity, which could obtain high mobility. Due to different process method, this lamella will be formed edge-on or face-on result. In my experiment, First of all, using chloroform to be a solvent with 0.1 weight percentages to fabricated OTFT can obtain better performance than xylene made, due to solubility and like dissolves like. Subsequently, we study the difference between regioregularity and molecular weight, two solutes were purchased from two different companies, dissolved them in the same solvent, chloroform. The result is that the difference of molecular weight is more importance than regioregularity. Moreover, fabricated by spin speed at 1500rpm and curing at 150℃ for 3 minutes can exhibit optimized result and in the meantime, observe that the oxygen concentration is less and less in the thin film analyzed by Auger electronic and ECSA surface elemental. Known a basic process procedure, there is a Schottky barrier between metal and organic material resulting less the current transport. However, S/D deposited with gold and an appropriated adhesion layer can reduce this barrier height, due to their closest work function and surface chemical property. In addition to previous results, to use surface treatment (HMDS) is a good idea to improve the packing and ordering for strong chain interaction and lead to exhibit high mobility and approximate to VG=0 of threshold voltage. In end of this thesis, we discuss about stability characteristics. Storage in the atmosphere, all of organic material will degrade as time goes by. This result is the same as P3HT polymer when it has been stored for one month. But the off current can be “healed” by annealing process at 180℃ for 30 minutes in furnace with N2 flow. Extra, low molecular weight comes to the same thing. Meanwhile, a device undergoing a serious process procedure can obtain a better performance, too. The on/off ratio is improved from 6x104 to 8~6x106; suspect that oxygen molecule can be driven off during annealing process. Unfortunately, the mobility is not to be improved and reduce down to 0.3x10-3 from 6x10-3 slightly and Vth is shifted to more negative, due to chain torsion. Last, constant voltage stress is applied to see how the drain current change. During stress, the drain increases with stress time going by when only drain electrode is applied at -60 voltages. There is no any gate current to be observed. We find out that the basic electrical result of the same device is changing after stress. A slight movement in off current, compared with just annealing one. So we suspect that the oxygen may diffuse into the film during stress.