SPM measurements in glovebox and local conductivity measurements on organic (semi)conductors.

SPM measurements in glovebox and local conductivity measurements on organic (semi)conductors.
A. Alexeev¹ and J. Loos²,
¹NT-MDT Co,
Russia ²Eindhoven University of Technology, The Netherlands.

	Fig. 1. SPM in glovebox.

Some of samples and SPM methods are sensitive to oxygen level or humidity and need controllable atmosphere. Different types of set-up can be used for such measurements: vacuum SPM (SOLVER-HV-MFM), hermetic cell with inert gas flow (NTEGRA-Vita) or glovebox with controllable atmosphere with SPM placed inside (Fig. 1). Commercially available gloveboxes are usually the large size containers filled by inert gas such as nitrogen or argon. Manipulations inside glovebox are executed by means of rubber gloves. The oxygen and water level inside glovebox can be less than 1 ppm. The SPM measurements in glove box have several unique advantages:

• Possibility of sample preparation inside glovebox without contact with air or transfer samples into the glovebox in hermetic capsule
• It is easy to exchange samples or cantilevers without contact with air.
• In-situ external influence (e.g. deformation or illumination of the sample) can be easily applied during measurements in inert atmosphere
• Additional equipment such as optical microscope can be installed inside glovebox without additional R&D work

The example of the importance of measurements inside glovebox is shown in Fig.2. The sample is the thin film of magnesium chloride deposited on the silicon wafer from solution by spin-coating inside glovebox. The left picture represents topography of the sample obtained after film deposition. The right picture shows drastic changes of the film morphology after short contact with air.

Fig. 2. The changes of the magnesium chloride film morphology after short contact with air: left picture was obtained just after film deposition; right image was obtained on the same sample after short exposure to the air.

The high resolution electrical measurements of organic conductors and semiconductors are of great importance since organic electronics have appeared on the market. Some of modern polymeric semiconductors are quickly oxidized in air. The local conductivity measurements with conductive cantilever in normal conditions also lead to the degradation of polymer. Measurements of local conductivity in a glovebox help to avoid chemical changes of the samples and allow obtaining reproducible and reliable data.

	Fig.3. Scheme of conductivity measurements on the organic electronic device.

The results shown below were obtained on the mixture of two amorphous semiconducting polymers. Both polymers are derivative of polyphenylenevinylene (PPV), one has hole conductivity (MDMO-PPV) and another one is electron conductor (PCNEPV). Such a mixture can be used as an active layer of solar cell. Both components degrade in normal conditions. The polymer blend was deposited on glass/ITO/PEDOT:PSS by spin-coating. The gold coated cantilever was used for measurements of the local conductivity. The ITO film with conductive layer of PEDOT:PSS was grounded during measurements and voltage was applied to the tip (Fig. 3).
Morphology of the polymer film is visible in Fig. 4a: one polymer forms domains inside another one. The thickness of an active layer is about 30 nm; height variations are significantly smaller than the film thickness. Conductivity measurements are performed in contact mode with conductive cantilever. The soft cantilevers (CSG10 or CSG01) should be used for conductivity measurements in order to avoid surface destruction. Noncontact NSG03 type of cantilevers is also suitable for measurements in contact mode on some of samples. It allows also nondestructive surface checking in semi-contact mode after conductivity measurements.
Fig. 4b and 4c show distribution of current on polymer film. When tip is positively biased the distribution of current correlates with topography (Fig. 4b). The regions with increased level of current (brighter regions on Fig. 4b) correspond to the MDMO-PPV matrix [1]. The distribution of current on matrix surface is nearly uniform. The negative bias on tip leads to significant changes of contrast: PCNEPV domains still have small current, while MDMO-PPV matrix shows nonuniform current distribution (Fig. 4c). Most probably, small current nonuniformities reflect a local organization of PEDOT:PSS, which is hole injector in this case.

a)		b)
c)

Fig. 4. Topography (a) and two current images (b and c) taken at opposite polarities of voltage: U=+8V (b) and U=-8V (c). The arrows indicate the same domains.

More complete information about electrical properties is obtained by means of I-V measurements at each point of scan. The array of such data can be obtained and analyzed by standard NT-MDT software. Fig. 5 shows results of such measurements executed at 128x128 points. Analysis of data reveals three types of I-V curves: one is typical for PCNEPV domains and another two are extracted from different places on MDMO-PPV matrix. It is visible that both I-V curves obtained on the matrix at positive bias are almost identical. At the same time the negative bias on tip leads to significant changes of I-V curves obtained at different places on the matrix. This difference appears as current variations on the matrix on Fig. 4c.
Standard SPM software supports also measurements of current-distance (I-z) curves [1]. It allows analyzing the dependence of current on load.

Fig. 5. I-V measurements at each point of scan (128x128 pixels): three types of I-V curves extracted from obtained array of data (top) and distribution of current at different voltage on tip (bottom).

Acknowledgements: Authors thank Dr. M.M. Koetse (TNO Industrial) and Dr. P. Thüne (Eindhoven University of Technology).

[1] A. Alexeev, J. Loos, M.M. Koetse, Ultramicroscopy 106 (2006) 191.