This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The study of dynamic processes was limited to the range of Micro seconds in a typical Scanning tunneling microscope .Now with the availability of ultra-short laser pulses combined use of scanning tunneling microscopy Ultra short pulse laser has enabled us to study ultrafast characteristics of a material in an excited state with ultimate spatial / temporal resolution, materials in excited state exhibit unique properties which are often in the order of femtoseconds. The resolution of this technique is limited only to the difference in the pulsing frequencies which is in the order of femtosecond 10^-15seconds. The excitation is provided by means of short pulse laser stimulation preferable of very high frequency .By studying the tunneling current contribution due to the laser stimulation it is possible to gain information about some of the surface and subsurface properties of the material with very high temporal and spatial resolution
Keywords - Scanning tunneling microscope, FemtoLaser stimulation, Ultrafast processes
Over the years atom manipulation was and its underlying principals have been discovered, and the atom-manipulation and study process has now become a reality.As early as 1959 Richard Feynman invited scientists to a new field of research with his speech "There is Plenty Of Room at the Bottom": the field was focused on studying materials at atomic level and to identify a mechanism by which they can be manipulated. By altering the atomic arrangement one could alter many properties and behavior characteristics of the material under study, many problems can be solved at an atomic level , thus giving us a permanent and more stable solution . The invention of techniques such as scanning tunneling microscope (STM) [2, 3] and the atomic force microscope (AFM) , have made this possible.
The STM works on the principle of Electron tunneling and is predominantly used to study the topography of a metallic material at an atomic scale. They can also be used to alter the atomic arrangement in order to imbibe required properties into the
material. Invented by Binnig, Rohrer, Gerber, and Weibel [2, 3] .Fig 1.1 describes the schematics of a STM. The STM is driven using 3 mutually perpendicular Piezo electric transducers, it consists of a very sharp tip that scans the surface of the material. When a Bias voltage is applied a tunneling current is setup, the value of the current is amplified and compared with a preset reference value. According to this difference the vertical position of the tip is varied (change in size of the piezo electric transducer due to difference in current) . Now when the tip is moved all over the surface in a systematic way the tip continuously varies due to variation in the tunneling current. the variation in tunneling current is in line with the variation in the distance between the tip and the surface . Therefore this vertical movement of the tip is the contour of the surface which is under scanning. This contour is usually displayed in grey scale in a computer screen or a CRT Fig 1.2a.Usually the image is also supported with a contour plot for further understanding. The STM's these days are computer controlled and have un paralleled accuracy levels it also allows to view the process as it happens. The facility to store and print the information is also available  , schematics in Fig 1.3.
Fig 1.1 -Scanning tunneling Microscope
We are now interested in studying the use of this microscope when combined with an amplified femtosecond laser stimulation .In this case we are more concerned in its ability to study the tunneling current rather than studying the topographical surface . When a region is energized by using an external source such a laser the material is excited at an atomic level , this influences the flow of charge carriers which in turn alters the tunneling current .The life span of these processes are in the order of Femtoseconds(Femtosecond is 1015 of a second that is one or 1/1000th of a picosecond ) . Along with the excitation a decay process is also inevitable the study of this process also enables use to study many sub - topographic properties such as carrier dynamics, electron absorption, Raman scattering of the material. The study of such ultra-fast dynamic processes is now possible by the combined use of STM and a femtosecond double pulse laser stimulation. The dynamic properties were found to have a direct relation with the laser frequency. Thus enabling us to study these Ultra-fast processes
This Concept of detecting the nonlinear response of the tunneling junction (due to laser stimulation) directly in the current ï¬‚owing through the contact was mentioned by Hamers and Cahill  The tunneling junction is excited by a two lasers with delay in the order of femtoseconds and the nonlinear tunneling current component that depends on their temporal separation can be studied.
The ultimate goals of this method is to monitor dynamic surface processes, surface topography , electronic surface states and defects with a much better spatial and temporal resolution when compared to the other available methods.
The concept behind achieving the higher resolution is the multi photon emmisivity of material when subjected to pulsed laser stimulation . In order to get past the electron bandwidth limitation we use repeated short pulse of the laser directly in the tunneling junction and study any non linear behaviour that is reflected in the tunneling current.Therefore the electronic response is now directly dependent on the delay between the pump and the probe laser pulses that are delayed in the order of femtoseconds Fig 1.4
This being the case now the lateral resolution of the STM is not due to the topography or diffraction, it I due to the non linear component of the tunneling current due to the laser stimulation. This paves the way to to study the visual interpretation of the electron dynamics when in an excited state.
Though we find this a straighforward case . Many limitations have to be overcome . To start off with, identifying the nonlinear component of the tunneling current is a challenge by itself as the current is fast and small , thermal expansion of the tip can cause a huge variation in the tunneling current , however this is overcome by choosing a laser frequency which is much faster than the tip thermal response time . But still the quasi static increase in temperature of the system has shown to increase the tip length in the range of 10-10 m. This by itself requires a close monitoring. The tip requires careful preparation in order to ensure that it scans only the required area . Small Variation in Laser Source also creates a challenge in getting accurate stimulation . Beyond all this the experiments have proved to give out information on charge carrier recombination, surface states and other atomic scale information [26,27]
The concept can be experimentally verified . A electrochemically etched tip was used to scan a clean GaAs sample . Output of an amplified TiSaphire Laser was used to stimulate the tunneling junction at a rate of 50hz . Michelson inferometer provided the required delay of 40fs .A short peak pattern in the tunneling current was noticed in line with the laser frequency. Pulse length was around 50fs in width . Due to the short rise and fall of this current there is no possibility that this is influenced by the dynamic distance controller of the STM (Z piezo) as it has a much greater response time . The current was also found not to depend on the tip to surface distance as the same current peaking was observed even when the tip was retracted. The governing factor for the current peak was noticed to be the actual laser spot position. Thus we can conclude and confirm that we are able to study the current component due to the laser stimulation which carries information at an atomic level .
Fig 1.5 shows the nonlinear variation of the of the current with respect to the incident laser flux, the graph also indicates that the scattering of values reduces with higher frequency of laser stimulation. Moreover the current was also controlled by the bias voltage a negative voltage reduced flow of current and Vice Versa. This proves that a tunneling component of current exists with the laster stimulation
Further to this in order to understand the decay characteristic of these excited electrons a pump and probe laser stimulation is required . Here two short laser pulses that are delayed at a femtosecond level illuminate the tunneling junction ( for the same GaAs material) and the current is studied against by varying the delay "The pump excites the electrons to a hot state and the probe takes them to a continuum state" .The below graph represent such an experiment
Note in fig 1.6 that the current reaches minima when delay is increased and slowly again reaches maxima. This gives us information on the Photo electron dynamica i.e "the current reduction can either mean that the pump laser depletes the charge carrier surface or that the charge carriers are moved to states which have less efficient emission coefficient". Furthermore, shorter pulse duration will increase the relative strength (between pump and probe) of the laser induced tunneling component making it a very desirable condition for our study  .
As we already saw there are many limitations and barriers such as small magnitude, non-linearity, ultrafast nature to study this nonlinear component of the tunneling current that carries the required information .These obstacles have created a number of features that are required for the setup. The three main divisions of the setup are
Femtosecond laser system
Scanning Tunneling Microscope
Fig 1.7 depicts the schematics of such a setup
The coupling quality in terms of harmonics and mechanics between the systems has a very critical role to play in ensuring the accuracy of our readings.
Femtosecond laser system and vibration isolation
In order to obtain the required result we require a laser which has the ability to pulse at frequency in line with the relaxation time of
the system which is in the femtosecond range thus we require a very quick pulsing mechanism . This requirement is translated into the setup by using a focusing lens to direct the laser towards the tunneling junction .The positional accuracy of the laser is very critical as we already saw that the tunneling current component is directly proportional to the laser position. "Usually focal length of 20 cm that is positioned outside the Ultra High Vacuum chamber. The angle of incidence of the laser beam with respect to the surface normal is about 60 - 80Â°. This position of the laser focus on the tunneling tip is controlled by a steering mirror that can be moved parallel to the tip axis using a translation stage. Movement of laser spot across the apex causes a diffraction pattern to appear, as we move it further down the tip apex this pattern changes . When the laser is focused at the junction the diffraction pattern is least influenced by the tip and by this we ensure that the laser is focused exactly in the junction A focus diameter of 50 mm is typically achieved. The size of the focus ~ 50 mm diameters is sufï¬ciently large to monitor the position of the laser focus on the tip or on the sample using a binocular microscope with large working distance".
The typical laser source we use is usually the direct output of a Ti: sapphire femtosecond oscillator or ampliï¬ed laser pulses after chirped pulse ampliï¬cation. Femtosecond oscillations is produced by a femtosecond oscillator capable of delivering ultra-short laser pulses of 40 fess duration at a wavelength of 800 nm ~80 MHz repetition rate, 4 NJ pulse energy. The ampliï¬ed laser pulses are then compressed and pulses of 60 fs duration and 2-3 mJ energy are obtained at a repetition rate of up to 300 kHz. "Second harmonic generation in a thin beta-barium borate ~BBO Crystal is used to generate short laser pulses at 400 nm using either the output of the femtosecond oscillator or the ampliï¬ed laser pulses. A Mach-Zehnder interferometer generates pump and probe pulses with variable time delay" .This describes a typical laser setup that is used for such experiments (all values are taken from the reference and are for purpose of explanation)
Scanning tunneling microscope
The UHV-STM for the discussed experiment has to full-ï¬ll several requirements. First, the excellent vibration isolation so that the focusing capability of the laser system is not hindered. A second requirement is that the STM should provide sufficient space to enable optical access for the laser which means that the STM should be able to measure the tunneling junction from a larger distance. A third important requirement is the geometry of the tunneling tip . The tip should have a very sharp apex so that it picks up only the required tunneling current although this can be guaranteed by the tip preparation methods , the slow change in tip geometry due to the thermal expansion over a period of continuous pulsing cannot be avoided . This change is dimension though will be useful while studying the topography of the specimen , it leads to error when we are interested in the nonlinear tunneling , Therefore our setup should have a facility to replace the tip and attain the same positional accuracy every time we do so . The STM therefore must allow a fast and reliable tip replacement procedure and in "situ tip characterization" .
The STM is highly sensitive to external disturbances and great care must be taken to keep vibrations away from the STM to ensure good resolution under laser operation. We use a STM that is coupled over a "Viton stack" ~see Fig. 1.8 this allows us to alter damping properties based on the external conditions. It acts as a low pass filter for external vibrations and cut off frequency Is about 100hz
Now this damping is sufficient when there is no laser operation. With the inclusion of the laser system we are faced with another problem. The laser system requires a cooling system to be run in order to keep the laser in a stable condition. This system creates vibration during laser operation and this hampers the measurement process. The cooling system is very powerful and the vibrations are of higher magnitude ( usually centered around 40hz). This is overcome by adding additional water reservoirs and extra-long hoses .To a large extent these vibrations are damped and thus enable us to have the required accuracy .
"Study Of Hot Electron tunnelling "
Introduction-"The development of semiconductor physics and devices has been progressing with the evolution of atomically controlled fabrication technologies that enable us to drastically alter the properties of the material "[18-20]. "However, till date the carrier dynamics in the materials have been analyzed only with techniques that provide spatially and/or temporally
Resolution only in the range of microseconds" . With the introduction of the above procedures we can now study the behavior of these carriers on a femtosecond scale .Fig 1.9 shows the schematics for such an experimental setup.The pump and probe pulse trains are generated by two Synchronized Ti:sapphire lasers (Mira and Chameleon, Coherent , central wavelength = 800â€‰nm, repetition rateâ€‰â€‰= 90â€‰MHz, pulse width = 140â€‰fs) they were guided to a pulse picker, which consisted of Pockels cell and polarizers. The delay time between the two pulse trains was continuously varied in the femto and nano scodne scale .The pulse train repetition rate was set at 1Mhz.The two pulse trains from the two laser sources were combined on to the same optical setup and guided into the STM scanning the specimen. The light spot, diameter of was measured to be around 10â€‰Î¼m and an average intensity of up to several mill watts was recorded . This was focused on a sample surface below the STM tip using lenses placed outside the UHV chamber . The tunneling current signal from the STM preamplifier with a bandwidth ofÂ ~10â€‰kHz was phase-sensitively detected using a lock-in amplifier so that only the required component was measured . In this application we are interested in studying the a Gallium Phosphide surface . GaP being a wide gap semiconductor turns out to be a very suitable material for our experiments. With an indirect gap of 2.27 eV and a direct band gap of 2.9 eV, therefore absorption of light at 800 nm ~1.55 eV is only possible via a two-photon
process (231.55 eV53.1 eV). For GaP, the intermediate states lie in the band gap, but GaP
is known to have a variety of surface states at these energies with lifetimes in the picosecond range. (all values are taken from the reference and are for purpose of explanation).
The tunnel process under illumination is inï¬‚uenced by the presence of excited electrons in the material and the strong optical ï¬eld in the form of the laser. These effects can increase the effective tunnel probability due to two different mechanisms: coherent photo assisted tunneling and tunneling of hot electrons. Therefore If the excitation of electrons occurs via a two-photon Process, the time-dependent current component due to the tunneling of these electrons carries information about the dynamics of the intermediate level. The life of the electron in each stage has information which previously could never be studied Now a modulated laser source in the form of Pump and probe pulses are used to illuminate the junction a bias voltage is applied and can be varied as well . On setting up a bias voltage the tunneling current is generated. The curve of the tunneling current will now contain the nonlinear variation due to the pulsing laser as well.
Fig 2.0 denotes the current fluctuations due to the Piezo scans . We are now able to see both the tunneling current due to the topography, usually denoted by the bigger peaks and the smaller peaks that denote the variation due to the laser emission. These peaks have a uniform height and thus are not caused by some instability in the junction .Therefore we can conclude by saying that these small negligible peaks induced by the laser stimulation represent the local properties of the Charge carriers  .
Band Structure investigation "
The technique was also used to study the band structure of a cleaved LT-GaAs/AlGaAs and AlGaAs/GaAs interfaces . The current was studied for the illuminated an non illuminated conditions as shown in Fig 2. By drafting the graphs between the bias voltage and the tunneling current we were able to analyze the charge carriers for each junction. Fig 2.1 denotes the various graphs that were drafted from the experiment
Fig 2 schematics of the experimental setup
Fig 2.1 output graphs of the experiment
Results - As already disclosed the results of this experiment is greatly useful in studying the semiconductor composition such as majority , minority carriers , impurities , concentration of electrons and holes . The flow of current against the bias voltage was studied . The relation between the current magnitude and the bias voltage direction could tell us what was the charge carries that was inducing the tunnelling current . For example we could notice from the graph we could see that for non-illumination that the tunnelling current does not flow unless a bias voltage much larger that the band gap is applied this indicates that no impurities are present in our sample. We could also see that the onset voltage is negative and positive for different layers . Negative onset indicates the presence of electrons and positive onset indicates the presence of holes as charge carriers. Further the variation of the current can also be studied in illuminated condition. It is usually found to be higher in magnitude and the onset voltage helps us to identify whether the layer has "n" type or "p" type behaviour. Further we can study the tunneling current with respect to decay time , this gives us a whole new set of information and avenues to experiment on the nature of flow of the charge carriers in the semiconductor , their decay time can also be studied at a femtosecond scale .Though this technique is not direct ,careful study of the information and comparison of results of other known methods give us a good understanding of the UHV STM applications .
The ultimate goal of an ultrafast laser assisted STM is to to monitor the electronic and structural surface dynamics on an atomic scale with femtosecond time resolution. The above experiments and analogies proves that it can be achieved. However this required careful identification of the nonlinear tunneling current contribution due to the laser stimulation. This current is dependent only on the frequency variation of this ultra-fast laser thus enabling us to reach the femtosecond resolution. Thus we can conclude that this new technique is very powerful and its uses are yet to be fully explored.