Result and discuss

Initially, there are so many P3HT precipitations in xylene solvent when it is dissolved to be a solution and the same situation is observed in chloroform with weight percentage higher than 1wt%. The solute of P3HT can be dissolved in chloroform easily, whereas xylene had poor solubility. Further, solvents and solutions will be gelled, although it is reported that both of them can dissolve P3HT solute. Two likelihood reasons for this are that “supersaturation”

and “like dissolves like”.

For supersaturation, this explanation is based on fundamental chemical. It defines that contains more of the dissolved material than could be dissolved by the solvent under normal circumstances, meaning each of solution has its supersaturation nearly. If the proportion of solute is lower a level, solvent could dissolve them easily. Thereagainst, if the proportion of solute is beyond, it would not easy to dissolve.

For like dissolves like, it is related to polarity. Polar compounds dissolve in polar compounds and non-polar dissolve in non-polar. The types of intermolecular forces dictate these results. Moreover solution is a homogeneous physical mixture of solvent and solute.

The solute particle sizes are ions, atoms, molecules or small combinations of these units.

Basically, polarity of material compounds are decided by polarity of molecules and polarity of molecules are responsible for its polarity of bonds. Let us start off the solvent. Chloroform (CHCl3) has 3 bonds with chlorine (two of them symmetrical; see Table 2-2), and one with hydrogen. Chlorine, like many of its neighbors on the right side of the periodic table, is very electronegative: it pulls electrons towards its nucleus, like so:

- 15 -

+

-Positive pole Negative pole

Thus, it is a polar bond. The two symmetrical chlorine-carbon bonds "balance out" this pulling, so that no net pull occurs. The remaining chlorine-carbon bond is not "balanced out" by the less-polar hydrogen-carbon bond, resulting in a net polarity, with the "-" end towards the chlorine atom. Xylene (Paraxylene, C8H10 (C6H4C2H6)) is low-polarity [2.9].

Although the side chain is less-polar (C is negative pole and H is positive pole), they are located at each side. According to this result, the polar of CH3 bonds balance out and its polarity is low.

At room temperature, xylene solution will gel within a few minutes. In addition, when devices were fabricated in xylene and exceeded 0.3wt%, the solution was not easy to pass through a 0.20 μm pore-size PTFE membrane, due to precipitation should block. About chloroform, it will dry out within several weeks, depending on low wt% ratio.

The ratio of weight percentage was defined as:

wt％ = (grams solute / grams solution) X 100％

(2-2)

There are two equations for calculating weight percentage, but their results are the same.

2.4.2 Electrical results

A set of transistors was fabricated with different weight percentages. Each new transistor set was measured and added to the plot in following figures. Although there is a vibration across these devices, the trend of measured result is stable fairly.

Figure 2-5 shows the transfer characteristic of P3HT transistors from different weight percentages and solvents. For P3HT in xylene, as you can seen in Fig. 2-5 (a), the result of 0.01wt% and 0.05wt% indicates that the data it is near a limit of measured machine, so a obtained curve may not be true, and there is no yield within 0.05wt% when measure. We

- 16 -

suspect that thin film is not to be formed so that obtained result is poor. In figure 2-6 (a), AFM surface morphology illustrates that the roughness of P3HT transistor in xylene with 0.01%wt is close to SiO2 one. For 0.1wt% in xylene and 0.01wt% in chloroform as shown in Fig. 2-6 (b), the surface morphology reaches around 3.5nm. It indicates that the roughness of exhibited performance is around 3 nm, but the roughness of xylene with 0.01wt% is below that. In addition to phase data, our think, thin film forming should be an issue.

Low weight percentage, 0.1wt% and 0.3wt%, the device turn on approximately at Vg=0.

In chloroform devices, as shown in Fig. 2-5(b), lower weight percentage exhibits the same condition, except 0.01wt% and 0.05wt%, and they exhibit a similar curve to 0.1wt%. Despite their result are closed to 0.1wt%, but it is not good enough for on current. As following P3HT in both solvent with weight percentage increasing, the turn-on voltage is shifted toward to positive values. In other words, the off-current is increasing when weight percentage boots.

Two possible reasons for this condition are that first, the precipitation and impurity more and more in active layer with high weight percentage could exhibit a defect, when the solvent reach its own supersaturation.

Second, weight percentage can be affected by the viscosity so that the thickness is an index for supporting this result. As you observed in figure 2-7, indeed, the thickness increases with weight percentage rising (the rest of part of thickness, P3HT in xylene and chloroform, please refer to Table 2-3). It could involve more the number of defects in the film.

Additionally, those defects can act as trap center, and transport carrier will be trapped during the device operating in accumulation region. Trapped carrier will be released by applied a reverse voltage.

About on-current, it increases with weight percentages booting clearly. Figure 2-8 shows the surface morphology, which scanning by AFM. P3HT transistor in xylene observed a clear line-like structure onto the SiO2 after spun. The line-like structure, named nanofibrillar, is

- 17 -

consisted of nano-rod. This nano-rod is relative to ligands, which is control agent. The component will affect growth of nanorod. Apparent result also observed with 2 weight percentage. We suspect that the P3HT thin film is a long-chain polymer with lamella layer structure. The carrier is transferring through pi bond, which located at above sigma bond, and also it is source of electrons cloud. With 2 weight percentage, there is a possible chance to form a clear and long line. In contract, less weight percentage, like 0.1wt%, cannot form that.

The result agrees from the data published by Yang et al. [2.10], who explains that long nonafibrils are composed of nanorods with width of ~22nm. He also suggests that all of long nanofibrils exists grain and grain boundaries. On the other hand, these grain and grain boundaries can be a trapping and releasing center so that affects the result of off current further.

Nanofibrillar provide a conductive path, and transfer carrier can go through it easily.

Also it improves a chance of hopping transfer. Surprisingly, this data has not agreed with P3HT transistor in chloroform, as shown in Fig. 2-9. Nanofibrils are not clearly and even short in sample of 2 weight percentage, compared with samples in xylene. In contrast, in low weight percentage sample, nanofibrillar is to be formed, due to its boiling points. Xylene solvent has high boiling point around 138~139℃, and about chloroform, it is around 60.5~61.5℃ (obtained from the MSDS of Aldrich-sigma company).

Many researches group [2.11] [2.12] reported that the elongated nanocrystals are strongly affected by solvent-evaporation rates and costing solvents. This is due to its low boiling point and rapid evaporation speed. For high weight percentage in our study, it is not enough time to interact and self-aggregation in molecule before P3HT dry out. This results a short nanofibrillar on SiO2 surface. For low weight percentage, it is probably that small number of molecule with semidry situation could self-organize to form the thermodynamic structure and efficiently to form nanofibrillar by nano-rod.

- 18 -

In fact, one more suspect is that higher weight percentage can obtain thicker film.

General speaking; a solvent will be dried out from outer (surface of a film). Self-organization inward film is on progressing when the surfaces of organic film dry out. So based on this reason, it may understand that on current increases with higher weight percentages.

Figure 2-10 shows the output characteristics of typical device with different weight percentage and solvents. The P3HT transistors show MOSFET-like characteristics, and they are obtained for negative bias. It indicates that P3HT behaves as p-type semiconductor. For small negative source-drain voltage (VSD) the FET operates in the linear region, and observed from result shows an ohmic contact. When the source-drain voltage increases, the gate field is no longer uniform, and accumulation area is formed. Beyond a certain VSD, the current becomes saturated. The shape of I-V curves are identical, but the magnitude is quite difference. Let the output characteristics of chloroform values minus xylene ones and the result shows in Fig. 2-11. It is indicated that the performance of P3HT field effect transistor in chloroform has a better result than that in xylene ones, and no matter what weight percentages, P3HT transistors dissolved in chloroform obtained the same result.