## TOF SIMS

Time-of-flight secondary ion mass spectrometry (TOF-SIMS) is a technique used primarily for the study of surfaces of solids. In simple terms secondary ion mass spectrometry is the mass spectrometry of ionized particles emitted when a surface is bombarded by energetic primary particles. Beyond positively and negatively charged ions, neutral particles are emitted. The ions are typically a mixture of atomic ions molecular fragment ions, cluster ions, and molecular ions (if sufficiently small). This allows for a detailed chemical analysis of the surface by providing a mass spectrum. The TOF-SIMS variant of the technique works by assessing the time-of-flight of the secondary ions and thereby achieving a high resolution of the mass of the ions. The usefulness of the method depends on the precision of the timeof-flight analyzers and on the properties of the primary ion used. With modern equipment differentiation of different chemicals and isotopes is standard.

TOF-SIMS has been applied to characterize polymer solar cells mainly to study degradation. Examples of this include; characterization of oxygen and water induced degradation, pinhole effects and more. Diffusion of indium into the PEDOT:PSS layer was demonstrated by Bulle-Lieuwma et al. using depth profiling TOF-SIMS.DOI:10.1016/S0169-4332(02)00756-0 Van Duren et al. have demonstrated that nanoscale phase separation occurs in active layers comprising MDMO-PPV and PCBM.DOI:10.1002/adfm.200305049 The phase separation was not observed up to 50 wt. % PCBM, but at 67 wt. % PCBM almost pure PCBM domains in a surrounding matrix of MDMO-PPV was observed. The data was based on AFM, TEM, and TOF-SIMS depth profiling. By standard solar cell characterization a connection with the phase separation and the device performance was established. By studying TOF-SIMS images Norrman et al. demonstrated that oxygen diffuses through pinholes in the aluminum electrode of a normal geometry device structure.DOI:10.1016/j.solmat.2006.04.009 Using 18O2 isotopic labeling in conjunction with TOF-SIMS Norrman et al. also demonstrated that oxygen-containing species were generated throughout the active layer. It was demonstrated that the oxygen came from the atmosphere and diffused through the aluminum electrode and into the device.DOI:10.1016/j.solmat.2005.03.004

As is evident from the examples of the use of TOF-SIMS for the study of organic solar cells; TOF-SIMS can be an extremely useful technique for elucidating information from devices.

Figure 1: Schematic of the TOF-SIMS system featuring primary ion source and time-of-flight tube.

### TOF-SIMS theory and principles

The basic operation of TOF-SIMS is that upon bombarded by energetic primary particles a collision cascade is initiated from the sample surface, see Figure 1 for a schematic of the TOF-SIMS operation. Since the ion source is pulsed, using short pulses of < 1 ns, accurate time-of-flight measurements can be conducted. This is done as the emitted secondary ions are extracted into the time-of-flight analyzer by applying a high voltage potential, V0, between the sample surface and the mass analyzer. Secondary ions travel through the time-of-flight analyzer with different velocities, depending on their mass to charge ratio. For each primary ion pulse, a full mass spectrum is obtained by measuring the arrival times of the secondary ions at the detector and performing a time-to-mass conversion. The generated secondary ions are electrostatically accelerated into a field-free drift region with a nominal kinetic energy of $$E_k=zV_0=\frac{1}{2}mv^2$$ where $m$ the mass of ion, $v$ the flight velocity of the ion, and $z$ is the ion charge. As ions with lower mass have higher flight velocity than ones with higher mass, they will reach the secondary-ion detector earlier. As a result, the mass separation is obtained in the flight time $t_{TOF}$ from the sample to the detector. The flight time is expressed by $$t_{TOF} = L \left( \frac{2zV_0}{m} \right)^{-\frac{1}{2}}$$ $a$ and $b$ are constants based upon the instrument parameters, and $m/z$ is the mass-to-charge ratio of the ion. The equation is valid as long as the initial ion velocity is zero.Cotter, R. J. Time-of-Flight Mass Spectrometry Basic Prinsicpes and Current State. ACS Symposium Series 1994, 549, 16-48. A calibration of the measurement is necessary as electronic delays in the time measurement system must be taken into account. During this calibration the constants $a$ and $b$ are extracted from a least square fit using known calibration peaks in the spectrum. This calibration must be preformed for all obtained spectra. The equation can alternatively be written as $$\frac{m}{z} = a'\cdot t_{TOF}^2+b'$$ It is hereby clear that the mass spectrometry is not in fact determining the mass, but rather the mass to charge ratio. The mass spectrum is drawn as a histogram of counts for each time interval converted into mass per charge.

### Mass spectrum analysis

Mass spectra have several distinct sets of peaks, which include the molecular ion (if detected), isotope peaks, fragmentation peaks, and meta stable peaks. Different types of ion sources result in different arrays of fragments produced from the original molecules. It is therefore important that care is put into the correct assignment of mass peaks in the spectrum. Usually the first strategy for identifying an unknown compound is to compare its experimental mass spectrum against a library of mass spectra. With knowledge of expected signals it is possible to assign mass markers to a range of the peaks. Mass spectrometers work in either negative or positive ion mode as a result of the value of the acceleration potential. It is very important to know whether the observed ions are negatively or positively charged. Many molecules/fragments/atoms are only observed in either negative or positive mode.

One major disadvantage of TOF-SIMS is that it is not a qualitative technique. This problem is caused by the matrix effect as secondary ion formation is strongly influenced by electron exchange processes between departing spices and the surface. Thus the electronic state of the surface is critical and the surrounding material becomes highly influential on the ion yield. To circumvent this problem complementary measurement techniques can be used, one example is X-ray photoelectron spectroscopy (XPS). TOF-SIMS data can empirically be calibrated against XPS data resulting in a transformation of qualitative TOF-SIMS data into quantitative data, which is especially useful when studying low levels of photo-oxidation in materials used in organic solar cells.

### Practical considerations

Being a destructive technique certain experiments are not possible using TOF-SIMS. It is not possible to take a working solar cell, degrade it, do a measurement and then degrade it again. After the analysis the solar cell will no longer function. Therefore experiments must be carefully planned.

Contaminants can be a major problem when conducting chemical analysis using surface sensitive (~1 nm) techniques such as TOF-SIMS. Silicone is a common contaminant on surfaces, and it is easily introduced by various materials such as oils, greases, heater transfer fluid, sealants, adhesives, surfactant, and medical devices. The typical silicone is polydimethylsiloxane (PDMS), which has a very low surface tension and thus preferentially segregates on the surface of samples. The silicone contaminant on the surface will induce strong signals and the mass spectra will be filled with peaks from the silicone rather than the real sample. Silicone results in the characteristic peaks including $m/z$ 28, 43, 73, 133, 147, 207, 221, and 281.Vickerman, J. C.; Briggs, D. Tof-SIMS: Surface Analysis by Mass Spectrometry; IM Publications, 2001. Likely sources for silicone are latex gloves as some latexgloves contain silicones. It is therefore extremely important to handle samples carefully and avoid contamination. A preferable option is to use polyethylene gloves that contain no additives.

Gaining access to the layers of interest is an important part of doing mass spectral analysis of solar cell devices. To achieve this two different approaches are available. The first is doing depth profiling and the second is physically removing the layers (if possible).

Figure 2: Schematic of the depth profiling procedure. An alternative ion source is used to remove material.

Depth profiling is achieved by introducing a constant ion sputter phase in between the pulsed mode acquisition phase in order to remove material. Successive removal of each layer in the multilayer device thus exposing the interfaces, which can then be analyzed. The process is then repeated until the desired depth has been reached. The sputter process is achieved via a secondary gun, see figure 2. This sputter mode depth profiling by TOF-SIMS allows monitoring of all species of interest simultaneously, and with high mass resolution. As a consequence of differences in the sputter efficiencies through different materials the depth profile is not normally displayed as a function of depth, but rather as a function of sputter time. It is not straightforward to perform a TOF-SIMS depth profile on multilayered thin-film devices. Problems include interlayer mixing caused by the sputter process, in which small amounts of the sputtered material will tend to be pushed further into the next layer. This complicates the analysis of diffusion phenomena in the sputter direction. Bulle-Lieuwma et al. have reported an increasing bottom crater roughness when starting from the aluminum top layer, resulting in a loss of depth resolution.DOI:10.1016/S0169-4332(02)00756-0 Another problem with TOF-SIMS depth profiling is the significant charge build-up in the sample surface caused by the sputter ion bombardment, which in many cases is too extensive for the electron bombardment to be able to fully compensate for the effect. Charge build-up will decrease the intensity of secondary ions resulting in loss of sensitivity. Despite the short comings depth profiling is a valuable tool allowing information to be obtained from the entire stack including the interfaces and the bulk.

The other option of gaining access to the buried interfaces is to physically remove material. This can be done in either of two ways; by delamination or by exploiting solubility of the layers. Since the layers comprised in a polymer solar cell is often solution processed, the organic layers can subsequently be re-dissolved and removed. This is done by gently swiping the surface with a cotton stick soaked in a solvent capable of dissolving the layer and without dissolving the next layer. The process is repeated until the layer is completely removed. Delamination is possible with encapsulated devices, since peeling the encapsulation off typically reveals a buried interface. The method can also be used on non encapsulated devices with tape substituted for the encapsulation material. When performing delamination it is critical to determine if the delamination occurred as expected. This can typically be tested by acquiring ion images on both exposed surfaces.

### Isotopic labeling

Since TOF-SIMS is not a quantitative technique it is desirable to have uniquely identifiable markers. One way of achieving this is to use an isotopically labeled atmosphere. This allows incorporation from the atmosphere to be monitored. Since the natural ratio of isotopes is well known it is possible to identify the amount introduced under the labeled atmosphere. For this an atmosphere chamber is used, see Figure 3.DOI:10.1016/j.solmat.2008.02.008 Typical isotopically labeled atmospheres consist of $^{18}O_2$ or $H_2^{18}O$ (often mixed with $N_2$ in order to simulate ambient conditions).

Figure 3: Atmosphere chamber used for controlled atmosphere degradation.

## Current weather

Temperature: 14.67 °C
Sample temp: 13.79 °C