Home

Open Access

1. Light source Absorbing material Detector Figure 2 8 Figure showing the three main parts of an absorption spectroscopy set up for gas samples a laser source a sample absorbing the laser beam and a detector When diode lasers are used in absorption spectroscopy it is referred to as TDLAS Tunable Diode Laser Absorption Spectroscopy By letting the in jection current to the diode laser have a saw tooth shape it is possible to repetively sweep over the absorption line and detect it in real time The wavelength in the diode laser needs to be narrow and operate in single mode to be able to detect the absorption lines Lasers have a finite line width and 19 20 Chapter 2 Theory it is of great importance that this width is smaller than the absorption line to detect it 5 2 3 1 Line shapes An absorption line always has a finite width How wide it is and the shape of the line depends on the sample temperature pressure and the surrounding materials A way to characterize the broadening of the line is through stating the Full Width at Half Mazimum FWHM 4 Natural line width Every absorption line has a natural line width Avn This is related to the lifetime of the state as indicated by Heisenberg s uncertainty principle The principle states that with an uncertainty in time At there is always an uncertainty in energy AF and thus frequency 4 AE At gt DN gt 2 4 The natural line width is under stan
2. band Ey p region ME n region a p region E neregion ee o an Figure 6 The band structure of a homojunction diode laser The left picture 1s without a voltage put over the junction and the right with an applied voltage The homojunction lasers have a major drawback they cannot work at room temperature This is partially due to losses from absorption in the junction Heterojunction lasers Losses from for example absorption of the photons in the cavity can be reduced in heterostructure diode lasers Heterojunction lasers are much more used nowadays than homojunction lasers because they can operate at room temperature and do not need to be cooled The laser used in this laboratory exercise is of the heterojunction type The heterojunction lasers have an active layer which is a semiconductor sandwiched between the two semiconductor layers with higher band gap en ergies as shown in figure 7 Figure 7 The band structure of heterojunction diode lasers Laser production In a diode laser the emitted light will first be spontaneously emitted and with the help of a gain medium and reflectance on the cleaved facets the photons will be mirrored and a so called cavity is created As the emitted radiation is able to bounce back and forth this will cause stimulated emission and the diode will start to lase i e emit light with the same wavelength and phase if inverted population is provided The waveleng
3. cally investigated by testing different settings from a minimum to a maximum under which the signal could be obtained The time constant of the lock in amplifier was investigated for an optimal setting which turned out to be 1 ms Above this value signals were broad ened and under this value signals were too noisy This time constant was used for all ramp frequencies The amplitude of the ramp selects the observed wavelength range With a larger amplitude the signal is less prominent than for smaller amplitude but more of the surrounding features can be seen The frequency of the ramp affects the signal if it gets too large then the signal is smeared out see Fig 6 3 This is due to the set time constant on the lock in amplifier and the sampling rate of the oscilloscope The time constant averages the signal during this time and hence decides the numbers of samples possible With a too large time constant or a too high ramp frequency there will be an insufficient amount of samples to provide an accurate signal The practical frequency limit of the ramp was found to be 20 Hz for the lock in signal see Fig 6 3 but normally ramp frequencies around 5 Hz were used 6 2 GASMAS set up 20 Hz 40 Hz 60 Hz Figure 6 3 The influence of the frequency and amplitude settings of the ramp generator on the output lock in signal The modulation amplitude seems to have a relationship with the width of the lock in signal see Fig 6 4 A
4. 2 4 2 Basic principles Modulation means that a high frequency sinusoidal signal is added to a car rier signal e g the ramp that scans over an absorption line see Fig 2 10 The modulation frequency is also sent as a reference to a frequency and phase sensitive lock in amplifier The output signal from the detector is fil tered by the lock in amplifier using the reference frequency and analyzed Through modulation the signal is moved to a detection band at higher fre quencies where the noise level is lower 13 A noise getting attenuated at higher frequency is the flicker noise also known as 1 f noise since it is approximately proportional to the inverse frequency of the signal In an electronical set up this noise stems from the resistors 14 Modulation with the use of a lock in amplifier could be explained as sitting in a noisv caf in Quito with Spanish speaking people around vou If someone 2 4 Modulation AN soomyv OM 23ms Ext FS 1 5V Figure 2 10 A ramp i e a direct signal with an arbitrary superpositioned modulation frequency including a zoom in on the modulation talks Swedish a couple of tables away you can probably hear that through the noise if you are a Swedish native speaker Your brain the lock in am plifier singles out the characteristics the signal at the specific modulation frequency the Swedish language and the information is retrieved Using modulation in absorption spectroscopy is als
5. Outside of these limits the temperature regulation mechanism for reaching the extreme temperatures was too slow to be functional Normally it is advised not to go below 10 C or above 50 C because of risk for water condensation or thermal degradation 8 3 3 Function generators 3 3 Function generators Two function generators are used one to ramp the current to the diode laser and another one for the modulation frequency In this report they are defined as the ramp generator and the modulation generator in order to differentiate between them The ramp generator Thurlby Thandar TG215 is used to sweep the injection current with a saw tooth signal sent to the diode laser The frequency limit of the generator for the set ups used depends on the lock in amplifier and the oscilloscope see Sect 6 2 1 The modulation generator Thurlby Thandar TG1010 is used to produce the high frequency modulation signal which is added to the slower ramp see Fig 2 10 It provides the reference signal for the lock in detection The upper frequency limit is set by the frequency range of the reference channel in the lock in amplifier The modulation frequency cannot be higher than half the reference channel frequency limit since the second harmonic 2f of the modulation signal is studied The lock in amplifier used has an upper reference channel frequency of 102 kHz 26 Therefore a modulation frequency higher than 51 kHz cannot be used when the second
6. The detector used in GASMAS is a very sensitive instrument a photomul tiplier tube PMT Hamamatsu R5070A The incoming light releases elec trons from a photosensitive material which are successively and substan tially amplified thus producing a measurable current Amplification occurs due to the presence of a high voltage difference between the anode and cath ode of the PMT which is supplied by an external high voltage generator A series of intermediate dynodes divide the full voltage and the amplification is performed in an avalanche through the dynode chain It is of high importance not to let the output current from the PMT be too high since this might destroy the PMT In order not to exceed the maximum the output current is measured over a resistance to create a voltage that can be measured with the oscilloscope The maximum output current from the PMT is 100 pA but it is only linear up to 10 yA Modulation In some applications the absorption signal is smaller than the surrounding noise To find these small signals modulation techniques together with a lock in amplifier can be used When a diode laser scans the wavelength over an absorption line with a ramped signal a direct signal from the detector is obtained by reading di rectly the output of the PMT In some cases if the absorption is strong enough it is possible to see the absorption signal in the direct signal see Fig 13 In other cases the absorption is small an
7. according to the fitted curve According to earlier investigations this relation should be quadratic 12 The reason for this discrepancy could be stemming from the large detection area used and the spaces between the polystyrene foam slabs 7 3 Drying balsa wood Moisture in wood is a problem in certain applications e g in residential houses Moisture is also a basic parameter of quality of the wood and is normally estimated as the weight of moisture in the wood compared to the wood When moisture affects the wood the air inside the wood is replaced by water 30 By measuring the decrease of oxygen it is possible to measure the increase of water inside the wood 7 3 1 Method A 8x8 cm piece of a 1 cm thick balsa wood was put inside of a bowl of water for 6 hours The wood piece was removed and the oxygen absorption signal was investigated for about 12 hours as the piece was drying 7 3 2 Results ap pe son Time constant 155 min Leq cm 0 200 400 600 800 1000 1200 Time min Figure 7 4 The measurement points made over time while drying a piece of balsa wood An exponential fit and its time constant are also shown 61 62 Chapter 7 Experimental work As expected the oxygen absorption signal increased with time as the water in the wood was replaced with air see Fig 7 4 By fitting an exponential curve to the measurement points the time constant could be estimated to 155 min The piece of wood felt dry wh
8. modulation spectroscopy since the frequencies are much smaller than the half width of the absorption line 13 So far GASMAS has only been used studying molecular oxygen embedded in scattering materials The narrow absorption lines studied in molecular oxygen belong to the so called A band see Fig 2 13 at wavelengths around 760 nm The absorption is due to transitions between vibrational and rota tional states 5 In theory the only thing that needs to be changed to study another gas is the wavelength hence the laser 6 The detector might also have to be changed to be sensitive in the specific wavelength range 4 H ES Intensity 8 7 6 5 4 3 2 1 0 E D Figure 2 13 The narrow absorption lines belonging to the A band of molecular oxygen 20 2 5 2 Normalization When dealing with absorption spectroscopy it is of great importance to nor malize the signals to be able to compare results In GASMAS normalization is calculated by taking the height of the lock in signal divided by the inter polated DC intensity of the direct signal at the location of the absorption signature see Fig 2 14 and Eqs 2 7 and 2 8 The normalized signal is referred to as the GASMAS signal and denoted GMS 15 S Sop 13 2 7 GMS 2 2 8 S dir 21 28 Chapter 2 Theory Figure 2 14 An illustration of a theoretical direct signal and a lock in signal as they appear on the oscilloscope screen The definit
9. we need to take certain precautions e You should under no circumstances ever look straight into the laser beam Use the IR card if you need to investigate the path of the laser beam but make sure the reflected light from the card is aimed downwards e Take off rings and wristwatches that might reflect the laser light They can cause hazardous reflections when working with the laser beam e Make it into a habit to always turn off the laser when not making experiments e There are laser protection goggles two pairs in the lab e Remember that when working with dimmed light your pupils are larger and more light can come inside So these recommendations should be considered with attention Thus it is advisable to work at a reasonable light level in the laboratory Photomultiplier tube 1 Turn off channel 2 on the oscilloscope 2 Make sure the 5 kQ external resistance is connected on the input of channel 1 3 Make sure the voltage from the power supply for the PMT is set to zero Turn on the laser and set the temperature and current values if they are not set to the given values 4 Raise the voltage for the PMT to achieve 10 uA according to your preparatory calculation but do not let it exceed this value What this corresponds to on the oscilloscope was calculated in the preparations A large noise will probably be seen on the screen depending on ampli fication etc 15 5 6 Remove the external resistance with
10. 2nL mA mc f for 0 The free spectral range i e the separation in frequency Af between 2 orders m and m 1 is then C Site any The lock in amplifier Noise can degrade or even completely hide the signal you want to study in a measurement A signal to noise ratio S N ratio is given to describe the situation If for example the S N ratio is 10 it means that the top of the measured signal is 10 times higher than the average background signal When the S N ratio becomes really small it is hard to separate the peaks of the signal from fluctuations of the noise the background signal Although there are measurement methods making it possible to improve the S N ratio and sometimes even detect signals completely hidden by noisc It is important in some way to reduce the background noise from the measurement for the useful signal to appear In the lab a function generator is used This generator superimposes a modu lation frequency upon the injection current Since the emitted light is directly dependent on the injection current the light will contain the modulation frequency as well The signal is then detected phase sensitively at the mod ulation frequency with the help of a so called lock in amplifier With lock in techniques a mixer also known as multiplier is used generating the sum and difference frequencies out of two incoming frequencies One in signal to the lock in amplifier the reference signal is the mod
11. A few other lasers operating around 49 C also detected these absorption lines Chapter 5 Technology and knowledge transfer The task of the project was not only to assemble the equipment but also to collaborate on the local level to build up a knowledge basis around it Ac tivities were started aimed to help inspiring students to get involved in the project Since GASMAS is a rather new technique it was found necessary to introduce it to researchers from other units at EPN discussing potential applications with them in their own fields of research and invite them to par ticipate in collaborative experimental studies During the project a Diploma student Ms Yolanda Angulo was involved full time This was an excellent opportunity to share the experience gathered by the authors in Lund during the preparation phase Ms Angulo collaborated in all the aspects of the installation and development learning all the aspects of the set up At institutes such as LTH the academic and scientific structure is orga nized in a certain way to foster the education of new students Graduate students carry on as part of their duties the propagation of knowledge on methods and instruments to newcomers This is an efficient way to keep a basis knowledge at home However at places like EPN this does not apply Undergraduate and Diploma students usually have to help others on their same level to learn As aid tools for this process the following activi
12. Wy ee al MESS as de oe Se ok a a a TAS DISCUSSION viedo cae wg ops hk ade Oe Mek a ce ap BR BO a To CU APDE ed beak AAA oe at ESE ol Method ess Shah id AO a a Me OR Hs eo TCS GS a da et ee A A fos DISCUSSO ees 2 Soe Be ee GM a Be Sc oe ee SS 6 Cutting a Danana sa d ich aka dd da ee Mosk Method aa fica e Bot toe e Ee ee ee de LOZ ROUS a ek era ech a e Be do ee R Os DISCUSSION oe 2 5 4 2 GS aa BR SE bet A Gi Mehod gt 66a EA E TS RENS wean ae we ein cca oe or eS we Acedia ee ee Giles DISCUSSION a2 ur ae Er a lw ate BE ee eee 7 8 Nitrogen exposed tree tomato 06 TSE Mood accu amp E abei Sek hh OOS amp amp Bie Une SE e Toe PEUS e Gd te me Bt E es ae es es a S e a DISCUSSION la Bk oS Ba amp a A 7 9 Volcanic rock overview o e TOL o A hee ee Ee eee de ee Pai T92 FROSUNGS oS rd ae eee A A we TOS DISCUSSION 2s a a ws ES aS SS wk amp SEs 8 Summary and conclusions 8 1 Results of the PTO EC c d ke 4 ad e de de A we ci he ee amp LL o ANITA 8 1 2 Technology transfer sale 0 Baty dd a a Bio Assembly Ls puna e a bee es 8 1 4 Experimental results dopado a ea A a BO SUTIL WOK esaa eaa ea ia AAA ss Acknowledgements Bibliography A Work responsibilities B GASMAS poster C Presentation abstract 63 67 68 69 82 83 87 89 91 O D Rubidium laboratory exercise E GASMAS laboratory exercise F Manual CONTENTS 93 95 97 Chapter 1 In
13. a saw tooth shape the wavelength will be repetively swept The use of diode lasers in absorption measurements is abbreviated TDLAS which stands for Tunable Diode Laser Absorption 9 V V Figure 13 A sweep in frequency over the absorption line at vo I states the recorded intensity at vy and Ip the intensity if no absorption would take place Spectroscopy The diode laser has to be operated with a narrow single mode profile to be able to detect the absorption lines The sample In GASMAS the free gas of molecular oxygen inside scattering materials is studied Scattering occurs when an incident beam interacts with a particle and the reemission of the energy or parts of it is in many directions This effect is a result of the emission from the oscillating electric charges forced by the alternating electrical fields This scattering process results in the path length the distance the photons have traveled being greater than the thickness of the sample see figure 14 Figure 14 The distance the light travels in a scattering sample is highly dependent on the scattering of the sample This scattering process complicates the use of Beer Lambert law since the actual trajectory is more difficult to correlate with the sample thickness A unit called equivalent mean path length Leg can be used to describe the signal This quantity is obtained by the so called standard addition method discussed further ahead 10 The detector
14. from the detector and lock in amplifier an oscilloscope is essential The set up uses a 200 MHz digital oscilloscope Tektronix TDS360 with a GPIB port This port can be used in order to analyze signals using a computer program such as LabVIEW A 100 MHz analogue oscilloscope Tektronix 2235 was used in parallel with the digital one for a short period of time Analogue oscilloscopes are better than digital oscilloscopes at showing fast signals superpositioned on slower signals 28 as is the case when an absorption signal is studied in a ramp It was discovered that small features in the direct signal like mode jumps and absorption signals were easier detected with the analogue oscilloscope Chapter 4 Preparations The two set ups require diode lasers operating at different wavelengths ac cording to the absorption wavelength of the target gas Rubidium has strong absorption lines and hence they are relatively easy to find Oxygen absorp tion lines studied in GASMAS however are weaker and need modulation techniques Because of this inequality different methods were used to find the absorption lines for the gases As described before there are DFB lasers that can be used in absorption spectroscopy With these rather expensive lasers one do not have to perform extensive testing of the lasers One of the key features of the project was low cost so testing mass produced diode lasers was the option chosen It was also important to sen
15. gas in a rubidium cell by using an absorption signal When the principle is understood it is easily comprehended how the absorption path could be a part of a respirator hose or a couple of kilometers of atmosphere Preparatory problems First read through this laboration manual and do the tasks below thereafter iF N At room temperature 20 C the vapor pressure of rubidium is about 2x 107 Pa How many atoms per m does the rubidium cell contain The distance between the mirrors in a Fabry P rot etalon is 10 90 cm The index of refraction is n 1 511 Determine the free spectral range the distance in frequency between the transmission maxima of the etalon A diode laser cavity works in the same way as a Fabry P rot interfer ometer The free spectral range is 130 GHz and the index of refraction in the resonator is 3 5 How long is the laser How do you change the frequency of a diode laser Practically there are two ways Which natural frequency width does the Ds line of rubidium have if the excited state has a lifetime of 26 0 ns Which Doppler width does this line have at 20 C and 120 C respectively In Fig 2 there is a Doppler free experimental diode laser recording of a spectrum from the Dg line of rubidium Calculate the frequency separations between the Rb peaks Put the Rb peak furthest to the left number 1 as the zero of the frequency scale The frequency distance between the p
16. is the absorption cross section at the frequency 1 the center of the line 5 It is usually absorbance that is measured in absorption spec troscopy experiments 2 1 Light propagation in matter Intensity x Length Figure 2 2 The absorption of a beam with the intensity Ip according to the Beer Lambert law All materials absorb differently depending on the physical state and compo sition Solids liquids and gases all have different absorbing features Gases have much sharper absorption lines than solids or liquids for which the atoms or molecules undergo complex interactions 1 2 1 3 Scattering Scattering occurs when an incident beam interacts with a particle and the reemission of the energy or parts of it is in many directions This effect is a result of the emission of the forced oscillating electric charges from the alternating electrical fields 3 The scattering process results in the path length the distance the photons travel being different from the thickness of the sample see Scattering in Fig 2 1 There are different scattering processes and which one of them that occurs depends on the material and the wavelength of the incident light beam If the particles causing the effect are small compared to the wavelength it is called either Rayleigh or Raman scattering If the particles are large in comparison with the wavelength it is called Mie scattering The Rayleigh and Mie scattering are elastic effects the
17. making a poster creat ing a website translating and modifying a laboratory exercise about rubid ium absorption spectroscopy creating and carrying out a laboratory exercise for GASMAS and finally creating a manual with the practical knowledge gained by the members of the project during the entire project Chapter 8 Summary and conclusions 8 1 3 Assembly After the arrival of the equipment it was assembled and tested The rubid ium set up was made functional but the signals found were weak and un reliable The GASMAS set up was assembled successfully and tests showed promising enough results to allow for live experiments 8 1 4 Experimental results Experiments were made on fruits balsa wood and polystyrene foam after verifying the signal and making an extensive standard addition procedure Standard addition Standard addition procedures were made to investigate the relation between the standard addition curve slope and the sensitivity setting of the lock in amplifier A set of standard addition curves were measured for six different sensitivity settings and the relation could be calculated The laser beam needed to be attenuated so three different samples were tested for this pur pose polystyrene foam a folded paper and an optical filter The optical filter proved to be the most suitable attenuator Polystyrene foam Measurements on polystyrene foam have been performed to investigate the behavior of scattering in this mat
18. project cost around 100 each 9 The diode laser used for the rubidium set up has to be able to scan either the D line at 794 7 nm or the Da line at 780 2 nm see Fig 4 2 23 The diode lasers tested and used in the Quito set up only range over the D line 32 Chapter 3 Equipment The diode lasers are physically attached to a laser head Thorlabs TCLDM9 This laser head also contains a TE cooler element which either cools or heats the diode laser to the constantly monitored temperature given by the tem perature controller 24 Figure 3 1 The electronical equipment used for GASMAS 1 Laser driver 2 Temperature controller 3 Modulation generator 4 Lock in amplifier 5 PMT voltage supply 6 Ramp generator 7 Oscilloscope 3 2 Laser controllers Two essential parts of the equipment to control the diode laser are the laser driver ThorLabs LDC202 and the temperature controller ThorLabs TED200 The laser driver provides the current needed to drive and tune the diode laser The temperature controller is used to determine the temperature and then steadily keep the temperature at this given limit This limit is im portant since a small change in temperature will shift the wavelength of the diode laser The controller is able to regulate the diode laser temperature from 40 C to 150 C 25 However for practical reasons in the experi ments with our set up the temperature was limited to values between 15 C and 55 C
19. the lock in amplifier and had an amplitude of about 20 mV Thus the signal to noise ratio was not acceptable This problem was solved by only using the external resistance initially to determine the maximum current but while analyzing the signals the external resistance was removed and the internal resistance of the oscilloscope 1 MQ was used When measuring the molecular oxygen in a sample it is desirable to avoid measuring ambient oxygen If the laser beam would go through open air on its way to the sample the signal would be significantly affected by the oxygen in the air causing an offset to the absorption signal In order to avoid this and for the overall convenience an optical fiber was used to couple the light from the diode laser to the sample When using a sensitive modulation technique the interference fringes ap pearing at certain optical interfaces are a problem of great magnitude Ev ery flat surface reflects some light which can be reflected again from a facing surface causing periodic intensity variations that overlaps the absorption signals 5 3 These fringes show up as a wave like feature on the lock in signal see Fig 6 2 To minimize the reflections the fiber ends were polished in angles but the system still contained fringes after this adjustment By introducing random vibrations into the whole system the different reflecting surfaces move and the fringes even out if the signal is averaged Figure 6 2 I
20. the concentration temperature and pressure of the sample The absorbance can be calculated by measuring and Jo see Fig 2 9 and using Eqs 2 2 and 2 3 To be able to determine the concentration the optical path length and the absorption coefficient need to be known In samples where scattering is not an issue the optical path length is the same as the geometrical length of the sample It is also possible to determine the concentration by comparing the signal with a signal obtained from a calibration sample with a known concentration 2 4 Modulation Modulation techniques are used to enhance the detection of small absorption signals Frequently phase sensitive detection is accomplished by the use of a lock in amplifier 21 22 Chapter 2 Theory V V Figure 2 9 A schematic showing a sweep in frequency over an absorption line at vo I is the recorded intensity at vo and Ip 1s the intensity if no absorption would take place 2 4 1 Advantages When a diode laser scans the wavelength over an absorption line with a ramped signal a direct signal from the detector is obtained In some cases absorption spectroscopy on rubidium for example it is possible to observe the absorption in the direct signal e g see Fig 2 9 In other applications the absorption signal is smaller than the surrounding noise To extract these small signals from the background noise modulation techniques together with a lock in amplifier can be used
21. the exchange of gases when nitrogen left the sample and oxygen entered 7 8 2 Results The results see Fig 7 10 indicate an increase in oxygen absorption The measurement points were fitted with an exponential curve and the time con stant was estimated to 25 min 7 8 3 Discussion The increase in oxygen absorption signal increased as expected Since the fruit was not manipulated in any other way than the change of gas environ ment it is suggested that the increase in oxygen absorption signal is due to the increase of molecular oxvgen in the sample The skin of the tomato functions as a membrane providing the fruit with an environment related to the gas environment When the environment is changed the gas content inside the fruit changes as well 7 9 Volcanic rock overview Time constant 25 min 0 50 100 150 200 250 300 350 Time f min Figure 7 10 The measurement points and an exponential fitted curve for the measurement of oxygen absorption in a tree tomato over time Before the measurement the tree tomato was immersed in nitrogen gas for about 5 hours 7 9 Volcanic rock overview There is a great volcanic activity in Ecuador with volcanoes erupting in the past few years Thus understanding the nature of volcanoes is a very important issue for the country of Ecuador and investigating the presence of gas in volcanic rocks might provide important geologic information In order to give an overview for furth
22. the sample to allow a gas measurement Oxygen is not the most important gas in these contexts so investigations to determine the concentration of SO in volcanic rocks should be a future challenge Our hope is that the cooperation with the Department of Geophysics will continue and bear fruit in the future Many different types of investigation have been thought of throughout the project and many of them were never realized due to priorities on other fields Hopefully these ideas will be investigated further in the future One of these ideas was to investigate wine bottles The cork of wine bot tles helps the wine to keep a slow and steady exchange of oxvgen with the environment much like the skin of the fruits investigated Some wines are 8 2 Future work 79 held for decades to get the perfect taste and aroma Maybe the GASMAS technique could be used to determine how the wine is doing during this pro cess Perhaps the oxygen within the wine or within the air gap could provide interesting information about the status of the wine Acknowledgments First of all we would like to thank Prof Sune Svanberg who made this project possible His great enthusiasm for physics has inspired us his knowledge about the field has helped us and his friendliness towards us has been greatly appreciated by us Linda Persson has been our practical supervisor who always took time to an Swer our numerous questions and inspire us when we felt down She has be
23. the separation between the transitions approximately 1 GHz 29 making it possible to see the hyperfine structure Figure 6 1 W shape signals of two absorption lines in rubidium and their corresponding lock in signal 6 2 GASMAS set up The same problem with not finding the absorption signals was experienced with the diode lasers for molecular oxygen Thus a new search for the absorption signals had to be done This was once again done with a long absorption path and with the use of a photo detector The laser output could however not be analyzed with a spectrometer or a wavelength meter This made it hard to know how far the wavelength of the emitted light of the diode laser was from the absorbing wavelengths or how far a mode jump or a multi mode behavior was from a certain absorption line After finally finding the absorption signal an optimization was made on the modulation signal and the lock in settings After the achievement of an optimized lock in signal the length dependency was once again tested this time with a PMT The output current from the 6 2 GASMAS set up PMT was measured over an external resistance 5 kQ and analyzed on the oscilloscope As mentioned before the current from the PMT should not exceed 10 yA which corresponds to 50 mV with the external resistance The direct signal turned out to be very noisy when the external resistance was used The noise was found to probably be inherent of the input ports of
24. with liquid or solid components such as for exam ple water which has broad absorption features The basic components in the GASMAS set up are a diode laser a photomultiplier tube and electronics for the signal detection including equipment for the use of modulation tech niques see Fig 3 Currently GASMAS is focused on measuring molecular oxygen which absorbs light at around 760 nm Thus this laboratory exercise is also focused on molecular oxygen Oscilloscope et a tN eS ene EN TR LWd Y ee Figure 3 Figure showing a schematic GASMAS setup Molecular oxygen Every molecule has in addition to electronic energy levels also vibrational and rotational energy levels During the laboratory exercise an absorption line resulting from the vibration and rotation energy levels in the so called A band of molecular oxygen will be studied see Fig 4 The A band consists of many narrow absorption lines around 760 nm with a typical width of GHz in atmospherical pressure ha a gt 8 7 6 5 4 3 2 1 0 Sy 2 Figure 4 The absorption lines in the A band of molecular oxygen Diode laser Diode lasers were introduced about 40 years ago Since then they have become the most important type of laser and a daily part of our life such as in CD players bar code readers and printers They are small and cheap but more important in absorption spectroscopy experiments is that they are possible to tune in wave
25. 00 mV 120 mV 140 mv 160 mV Figure 6 4 The influence of the frequency and amplitude settings on the modulation generator on the output lock in signal 6 2 GASMAS set up Ambient light Ambient light was detected by the photo detector and caused a noise signal detectable on the direct signal Different types of indirect illumination like sunlight caused different offsets to the signal Light from light bulbs or flu orescent lamps caused a 60 Hz noise to the signal and an offset As expected from the modulation detection theory this signal did not affect the lock in signal as can be seen in Fig 6 5 The noise from the electrical lights in the direct signal was of course easy to get rid of by turning off the lamps and cover the optical table Turning off lamps was however not always possible because the laboratory was initially shared with an office Later a wall was built to separate these two rooms sub stantially removing this noise source This wall together with black painting of the windows also removed most of the offset caused by sunlight a Pl AN PA i y F n aN yt tjata o A 4 A l Y 4 AAD GE qn lg i ij Ki ES ray A wil y NS i 1 sf Cag sada at ssn keh F LlV Figure 6 5 Normal signal left and the same signal with a 60 Hz optically coupled noise source right The lock in signal is not affected by the noise Impact vibration The optics was based on an 60 x 45 cm opti
26. 1 Introduction A res AAN la Zo RUIDOSO dd e deh ae A ia die Gos we Teg AOL a ds Aids Bote Behe We ro a ee da 1 4 Achievements 0 0 0 0 0 02 200200004 US TOUGH 3 bea bl ee BU aa ke amp ee es aS 2 Theory 2 1 Light propagation in matter DAA RENEGON A e sa Base Bao BES SG Bs a Did PeOSOLOLION a ac oe a ra del de eo N 2A OCINE x 6 60 eg ha ee bey He Se ee we 2 2 The diode laser hm e aba a weds SLA et TION 4 40 a wk fee Beat ee BA Se a RN 22 2 NO VALLES A date kee Berd Si doe Boe oe ew amp 22 BASIC p incipl s Led eg ce we Be Bo So a aa 2 2 4 Optical properties vs 2 4 aa ea da Ge doe 22 Mode VOIDS a aios oa a e e HK Powe Bee a AS DaO Tima der ad re 2 3 Absorption spectroscopy e 23L Lipeshapes dba e da da he A DO ADOS ra Aa a A 2 NOOO LON aso 5 8 eased oh a ad AR A e OAs Advantages sra DE da a WS DAD Basic PDOACIPDIES rl a a 243 LOCK amphier ena od He et De MO E Zal BSGI lt 2k e ae wale a a LE Boe A AS De NORMA ALO Hi Feats tad tee eh 253 3 Standard addition s e s s ao e wn ok a 11 11 12 12 13 14 14 14 15 17 18 18 19 20 21 21 22 22 23 25 25 Zt CONTENTS Deo RESC ONS e 6 6 x a ik S74 a a a A A 29 Equipment 31 Bl DOTTI GS se Sis eee awe Hh Woe Ms Eek We FA ew 31 dle ASCE COMULLOLEES ca a we ee re Se A ee a O SE 32 3 3 Function generators 4 ar oki era ess He RS 33 Ds DEO aki aria ce eo ieee eae a eh he a A EN 33 SAT Photo
27. AS could give some information on their inner content of molecular oxvgen and their ability to be tested with the GASMAS technique This information could be useful not only to give an idea about the limits of the svstem but also to provide an idea about which fruits and vegetables that can be investigated in the future The fruits and vegetables chosen are all grown in Ecuador 7 4 Fruit and vegetable overview Figure 7 5 The fruits and vegetables tested Top row from left papaya bananas tree tomatoes potato Hawaiian papaya Second row coconut av ocado passion fruit physalis horitos cassava Third row pitaya apple guayaba naranjilla taxo granadilla Bottom row aloe vera 7 4 1 Method First different kinds of fruits and vegetables see Fig 7 5 were tested to see if useful oxygen absorption signals were obtained Some of the fruits that gave a molecular oxygen signal were further analyzed which is described in the following sections 7 4 2 Results Table 7 2 shows that the majority of the fruit gave a descent or good sig nal Exceptions were the coconut cassava and avocado which completely blocked out the laser light making it impossible to determine the oxygen presence in these fruits and vegetables In the Hawaiian papaya the laser beam penetrated the fruit but still no absorption signal was obtained 7 4 3 Discussion The results show that banana tree tomato passion fruit physalis horito apple naranj
28. B M J Muthoka Diode laser absorption spectroscopy for teaching undergraduate physics Africon 2 1247 1252 1999 BIBLIOGRAPHY 24 ThorLabs Operation manual TCLDM9 aug 2004 25 ThorLabs Operation Manual Thermoelectric Temperature Controller TED 200 jun 2003 26 Stanford Research Systems Operating Manual Model SR810 DSP Lock in amplifier 2000 27 W R Leo Techniques for Nuclear and Particle Physics Experiments 2nd ed Springer Verlag Berlin Heidelberg 1994 28 P Carlsson S Johansson Modern Elektronisk M tteknik 1st ed Liber Eskilstuna 2000 29 G Belin L Holmgren S Svanberg Hyperfine interaction Zeeman and Stark effects for excited states in rubidium Physica Scripta 13 351 362 1976 30 V P Negodiaev E V Sypin E S Povernov The monitoring system of wood moisture and air temperature in the drying cell of wood 5th International Siberian Workshop and Tutorial IV 225 228 2004 Appendix A Work responsibilities The project was performed by one student at the LTH Engineering Physics branch Marta Cassel Engquist and one from the Electrical Engineering branch Christoffer Bj rkwall This made marks in the way the work re sponsibilities were divided Below follows an approximate division of work for these two project participants The preparatory work took place in Sweden and includes preparational ex periments and arranging for shipping the equipment Administrative work includ
29. Ch 4 describes the preparations in Sweden Ch 5 the informational activities performed when the authors arrived to Quito Ch 6 the assembly and Ch 7 the experimental part The report ends with a summary and a conclusion Ch 8 including a brief discussion of future work Chapter 2 Theory To understand absorption spectroscopy knowledge of how light interacts with matter is needed Useful tools in absorption spectroscopy contexts are the diode lasers Many times in absorption spectroscopy the signals obtained are very small and a way to increase the sensitivity is to use modulation techniques which for example the GASMAS technique uses 2 1 Light propagation in matter When an incident beam of light strikes a material different interactions take place In general there are three possibilities the light can be reflected absorbed and or scattered see Fig 2 1 Depending on the specific material one or two of these actions dominate over the other REFLECTION ABSORPTION SCATTERING lt i 7 Ld So en Eee TR ENEN A A A A Se eek See eee A EEEE A ees See f Figure 2 1 The three possible interactions between light and matter reflec tion absorption and scattering 12 Chapter 2 Theory 2 1 1 Reflection A light beam that hits a surface between two media will be partially reflected due to the change in refractive indices n The reflectance R is calculated as na Nn N2 Ma where n is the refrac
30. Establishing a Diode Laser Absorption Spectroscopy Laboratory in Quito Ecuador Master s Thesis _ by Christoffer Bjorkwall and Marta Cassel Engquist Lund Reports on Atomic Physics LRAP 347 Department of Physics Lund Institute of Technology Lund September 2005 Abstract Diode laser absorption spectroscopy has many useful applications and has the advantage that the equipment in most cases is small cheap and easy to handle A laboratory in this field of research and education has been established at the Department of Physics at Escuela Polit cnica Nacional in Quito Ecuador as a Master s project The laboratory equipment was sponsored by the International Science Pro gramme in Uppsala Sweden It includes two systems one for studying rubid ium absorption and one utilizing the GASMAS technique GAs in Scattering Media Absorption Spectroscopy This technique has been used for in situ studies of free molecular oxygen embedded inside scattering media prop erties unique for this technique The GASMAS technique was introduced in 2001 at the Division of Atomic Physics Lund Institute of Technology Sweden The scope of the project was to prepare the equipment for transportation transfer technology and knowledge about it on site in Quito assemble the set ups and finally make experiments on topics of high potential for Ecuador Measurements on fruits polystyrene foam volcanic rocks and balsa wood were performed Contents
31. Mode jumps constitute a hazzle in absorption spectroscopy since they limit the possibility to tune the wavelength They severely limit the wavelengths possible to reach with a particular diode laser 8 NN Mode jump ie Figure 2 7 Mode jumps making discrete jumps in wavelengths when the cur rent or the temperature to the diode laser is changed 2 2 6 Tunability The great advantages of diode lasers for spectroscopy overshadow the disad vantages Diode lasers have high spectral purity high wavelength stability great modulation capabilities and most important of all tunability The possibility to tune the wavelength of the laser is what makes diode laser spectroscopy possible and simple 8 The wavelength output from a diode laser is dependent on both tempera ture and injection current There is a temperature dependency of the band gap and by varying the current the gain curve and thus the wavelength is changed The refractive index of the band gap is also temperature depen dent and can be altered by directly changing the temperature of the diode Thus it is possible to use both temperature and injection current as tools to change the output wavelength of a diode laser The methods differ regarding the wavelength shift produced A band gap temperature shift makes a dif ference of about 0 25 nm C and a change in refractive index of about 0 06 nm C 8 The relation between current and wavelength is in the order of 107
32. age Stop After R S Rulton Figure 1 A mode jump an absorption signal and the saw tooth shaped ramp PMT current with a zero 4 In figure 16 it is possible to see the signal obtained and the 2nd deriva tive of an absorption signal and a mode jump Make a sketch of how you think the 1st derivative would look like for the examples given 2 5 r seh mee ane eaan meara Coe mentera me nase A een ara an ee A A A enan ee e ma ch e m e l 2 3455 e i Ta i z 1 y 0 3227x 0 1886 i R 0 9997 os 0 PEA IN A A se A A peas cot A A A eee etic a a AA 0 1 2 3 4 5 8 7 Added distance cm Figure 2 Measurement plot for a standard addition measurement 5 Assume a sample has been studied and data according to Fig 2 was ob tained What would be the equivalent mean path length of the sample according to this function Use centimeters 6 When you perform the standard addition procedure What will theo retically be the difference between for example if you have a sample containing gas or if you have nothing but an optical filter The equiv alent mean path length The slope of the standard addition curve GASMAS GASMAS GAs in Scattering Media Absorption Spectroscopy is a tech nique that estimates the gas content inside scattering materials such as polystyrene foam fruits and the human body The GASMAS method uses the fact that gases specifically absorbs light with a very sharp and distinct wavelength in comparison
33. apolated line The equivalent mean path length Leq is dependent on both the concentra tion of molecular oxygen and the scattering coefficient of the sample In a highly scattering sample the light will travel a longer distance along more complicated pathways and hence there will be larger probability to find molecular oxygen The real concentration of molecular oxygen in the sam ple Csm can then be related to the concentration in free air by using Leg according to CairLeg Com or 4 where Lsm is the actual optical path length traveled by the light inside the scattering sample One needs then to know Lsm to determine Csm Or vice versa Laboratory exercises Laser safety considerations The diode lasers are small and have a relatively low output power but the laser beam can still be harmful to your eyes The diode laser used in this lab is of class IIIb which means that it is potentially dangerous to your eyes by direct incidence or by diffuse or reflected incidence lasting longer than 10 seconds The wavelength of the laser is about 760 nm which is a wavelength 14 where the human eye has low detection ability But even if the laser beam seems to be really weak it might still be powerful There is an IR card in the lab With this detection device it is possible to see laser beams at infrared wavelengths The diode laser we are using has an output power of about 5 mW This power is enough to damage the human eye and thus
34. bability a strong signal is still expected 4 Since the exact wavelengths of the absorption lines were known a wavelength meter was used to measure the wavelength during the signal search to simplify locating the absorption signals Most of the diode lasers tested had a wavelength of about 785 nm at 25 C in accordance with the data sheets Thus it was necessary to cool the laser 39 40 Chapter 4 Preparations 5 Paz D2 780 2 nm D 794 7 nm Figure 4 2 A selection of energy levels for rubidium 87 The transitions called D and D lines are indicated by arrows 23 to around 10 C to reach 780 2 nm or heat it to around 45 50 C to reach 794 7 nm It was however not possible to cool the lasers sufficiently to emit light around 780 nm due to restrictions in the temperature controller During the investigations very strong absorption signals were achieved Thus neither the lock in amplifier nor the spectrometer needed to be used The absorption signals were observed directly as an oscilloscope representation It was easy to verify that a signal really was an absorption signal and not a mode jump simply be removing the rubidium cell and see if the signal disappeared or heating the cell to observe the increase of the signal Each laser of the 18 examined was tested between approximately 13 C and 50 C with this method At an acceptable diode temperature 42 C two clear and distinct Da absorption lines were found
35. been checked and verified and the more complicated set up for GAS MAS could be assembled 6 1 Rubidium set up The lasers suitable for rubidium absorption signals found in Sweden turned out not to be so suitable after arriving in Quito Eventually the sought ab sorption signals were found at a ten degrees higher diode temperature than in Sweden and with a significant difference in appearance The system also showed instability a signal could suddenly disappear when no variables were changed After this the signal took a long time to retrieve and had again shifted in temperature Parts of these problems were probably due to the laser not working in a single mode To ascertain that the signal was not a spurious one the gas cell was either removed to see the disappearance of the absorption feature from the signal or heated to observe the increase of 46 Chapter 6 Assembly the absorption signal When the temperature is risen the vapor pressure increases resulting in more rubidium atoms being released from the metal deposit on the cell walls and hence an enlargement of the absorption signal can be seen The lock in amplifier was connected to optimize the detec tion and to practice working with the lock in amplifier with a well defined absorption signal The absorption signals showed a W form see Fig 6 1 This was due to the hyperfine structure being visible The Doppler broadening which is 0 5 GHz at this wavelength is smaller than
36. cal table which was put on a common wooden table This rather light weight construction made the set up easily affected by impulse like vibrations propagating from the sur roundings into the signals It was discovered that this was caused by the laser beam alignment moving along with the vibrations and thus different light intensities were measured at the photo detector This type of disturbance was somewhat reduced by putting the optical table upon a 10 cm thick bed of soft isolating protection foam 32 Chapter 6 Assembly Bad electrical ground Having a good electrical ground connection is very important when dealing with scientific equipment This was a concern at EPN and the set up was connected to the best possible ground point A connection to a supposedly better installation in a nearby laboratory with a long cable did not improve the situation but worsened the quality of the signals Probably the cable that connected the two ground points functioned as an antenna and thus was adding high frequency noise to the system 63 LabVIEW A computer program was developed using LabVIEW by National Instru ments to simplify the measurements and make data analysis faster This program made it possible to transfer the data from the oscilloscope directly into a computer The functions of the program include the ability to retrieve data from the oscilloscope memory and make automatic calculations The program calculates the GMS value from t
37. ce the absorbing sample and the detector What makes GASMAS unique is that it permits in situ measurements of free gas inside a scattering solid or liquid The GASMAS method can also give in formation about the pressure temperature internal structure and diffusion characteristics of the investigated material 12 The technique has since its first appearance in 2001 been used to investigate a wide range of applications and subjects such as polystyrene foam 12 gas exchange in fruits 17 18 wood 6 packaging plastics 18 and diagnostic measurements on human sinuses 19 2 5 1 Basic principles In the GASMAS set up the light from a diode laser with a sharp spectral output is sent into a scattering medium through an optical fiber see Fig 2 12 The light is scattered in the sample and the pores of the target gas absorb the light at its specific wavelength This results in a signature in the output signal This signal is detected with a photomultiplier tube PMT whose output signal is analyzed with modulation techniques For GASMAS two different measurement geometries are possible transmission and reflec tion through backscattering 1 18 20 26 Chapter 2 Theory Laser Driver mod T Diode Figure 2 12 Schematic picture of the GASMAS set up What enables the GASMAS method to sort out the absorption line from the free gas embedded in the scattering material is the line characteristic Gases have ma
38. currents were noted These results were extrapolated to calculate the threshold current and the maximum current values for each diode laser see Fig 4 1 P max Figure 4 1 Diagram showing how to calculate Ima and lin The two stars signify two arbitrarily chosen measurement points An extrapolated line dot ted from these points 1s shown together with a line indicating the empirically estimated dependency of the power on the current Signal search Molecular oxygen absorbs weakly at 760 nm so to find a laser suitable for oxygen spectroscopy a long absorption path was used This means letting the laser beam pass through as much air as possible gt 10 m in order to strengthen the absorption signal The technique was used to find possible absorption signals in the direct signal see Fig 2 9 4 2 Finding suitable lasers for rubidium To investigate these possible absorption signals a spectrometer was used The aim was to clarify if the signals were referable to absorption mode jumps or multi mode operation The spectrometer also showed the range around a certain wavelength where the diode laser was operating in a single mode This is a requirement for working around a specific oxygen absorption line The investigation was continued using a lock in amplifier The settings of the lock in amplifier were adjusted until a nice absorption signal was obtained With the help of the lock in amplifier additional investigations could b
39. d lasers to Ecuador that did not need to be tested there The facilities and equipment at EPN are not as suitable for exploring and finding the correct lasers as at LTH Therefore these preparations took place in Sweden The preparations lasted for about six weeks 4 1 Finding suitable lasers for oxygen Molecular oxygen has many rovibrational absorption lines around 760 nm in the A band see Fig 2 13 In order to search and find these lines with a tunable diode laser the maximum current values for different temperatures first needed to be determined not to destroy the laser When knowing the current values for which the laser could be operated safely the search for the molecular oxvgen absorption signals could be done 38 Chapter 4 Preparations Determining the maximum currents The maximum current Imaz and threshold current In had to be determined for different temperatures The threshold current is the injection current where the diode starts to lase and the maximum current is the injection current that gives the maximum power allowed Between 10 C and 45 C Imaz and the J were determined for every 5 C increment An extrapolation procedure was used to determine the threshold current for each laser at each setting point A power meter placed at about 5 cm distance from the laser was used to measure the output power of the diode laser For two arbitrarily output powers between lasing and the maximum allowed the injection
40. d modulation techniques need to be used The modulation techniques enable to measure a signal which is only at an order of 1074 1078 of the direct signal Modulation means that a high frequency sinusoidal signal is added to a car rier signal e g the ramp that scans over an absorption line see figure 15 The modulation frequency is also sent as a reference to a frequency and phase sensitive lock in amplifier The output signal from the detector is filtered by the lock in amplifier using the reference frequency and analyzed Through modulation the signal is moved to a detection band at higher frequencies where the noise level is lower according to noise theory Modulation in absorption spectroscopy is also referred to as derivative spec troscopy because the modulation signal gets the form of the derivative if the LI Figure 15 Ramp with a modulated high frequency sinus function including a zoom in on the modulation modulation is small Figure 16 shows the direct signal for a mode jump and an absorption signal and their resulting lock in signals Absorption signal Mode jump ESES Figure 16 A comparison between an absorption signal and a mode jump in the direct signal together with their corresponding lock in signals The lock in signal corresponds to the second derivative of the direct signal Direct signal Lock in signal Normalization When dealing with absorption spectroscopy it is of great importance to nor
41. dard conditions only at an order of 0 1 100 MHz 5 Doppler broadening At low pressure under 10 Torr the thermal motion of the atoms or molecules dominates over the broadening of the natural lines This is called the Doppler broadening Avp and is dependent of the temperature T and molecular mass M of the sample PT Avp const vo Vi 2 5 The Doppler broadened line has a Gaussian line shape 4 This broadening is about 1 GHz in the visible region and thus strongly dominates over the natural line width 2 4 Modulation Pressure broadening Collision effects dominate over other broadening effects at atmospheric pres sure producing a Lorentzian shape on the absorption line profile At stan dard conditions atmospheric pressure and ambient temperature the pressure broadening is about 3 GHz and thus dominates over the Doppler broaden ing and the natural line width 5 The collisions shorten the lifetime of the excited state because of deexcitation 4 The pressure broadening Av is dependent of the pressure according to Avy gt Y Pi 2 6 where p is the partial pressure and y is the partial pressure broadening coefficient 5 At intermediate pressures 10 100 Torr the resulting profile is a convolution of a Gaussian and a Lorentzian profile This is called a Voigt profile and has to be computed numerically 5 2 3 2 Analysis The properties which can be achieved through absorption spectroscopy are
42. detecto ee a ae Mh ord be Boh es 34 3 4 2 Photomultiplier tube ae eA ee oe 34 jo bOCKeIM amm OUNCE q aaaea E SB weer A Oe e aL 34 3 0 OSCIOSCODEG amp e ds 6 5 do AAS Oe ee SL a 35 Preparations 37 4 1 Finding suitable lasers for oxygen 37 4 2 Finding suitable lasers for rubidium 39 Technology and knowledge transfer Al Dele POSO ee ls Ge eared Gets Sot ee Oe ee ge 42 D2 Presem a sc ang do oe AR a ok We A 42 5 3 Laboratory exercises aa Aa aoe ee MA ee 43 DA Wan das oh deve aaa sa a 43 Ooo WEDO seston ds e a e a do a iaa 43 Assembly 45 6 1 Rubidium set up ase a de Ble ee Boge A iS Bones 45 0 2 GASMAS SCID daa BEV eee ow eh we be ee eS 46 6 2 1 Optimizing the parameters 48 6 2 2 Situation specific noise 2 o oo a es ee 49 Co Os oe te te er e a aar es a Ge a aa 52 Experimental work 55 TE Sardan adaton Ls a A ae BE Gd aa a SS 55 T 1 1 Sample dependency ze bid a a oe Ae HA Od 56 7 1 2 Lock in amplifier sensitivity setting dependency 57 2 PONS Vrone TOAN el Bok wee Ge Soke a Bo A 38 al Dieren widths 3 8 v4 wis 4a we Sa be ede oS 38 7 2 2 Different thickness bio po y dl ew arde da 60 Lo Denn basa WOO dy Dia da ee a de Be ee A 61 A A o A 61 oz CES MES a oe Ge E Awe O Ra amp a 61 133 DISCUSI N ene amp 1 ese 6b e a e eS 62 CONTENTS 7 4 Fruit and vegetable overview 0 0 0 0084 CET A oo aot oy elke ge te ae he ah es es en ee ne oe
43. dia the sample The laser ight is guided to the scattering sample by using a fibre When light has passed the scattering medium the distance the photon has traveled is always equal or larger than the thickness of the sample depending on the sample s scattering properties PMT the defector The photomultiplier is a very sensible detector that measures the light that has passed the scattering media Lock in amplifier The lock in amplifier is a very essential tool to detect and measure very small signals even though the noise and the background are several orders larger in The body contains many gases and the GASMAS technique can be used to obtain important diagnostic information A certain gas concentration can be an indicator of an Miness or homeostatic imbalance Domestic industries ike the flower banana and alpaca industry can i have great use of GASMAS magnitude technique since they all deal with biologic materials Acknowledgments a The GASMAS equipment was donated to the Departamento de Fisica in Escuela Politecnica Nacional Quito from the Intemational Science Program Uppsala There are two types of detection geometries that can be considered when Sweden in collaboration with the Department of Atomic Physics at Lund University performing GASMAS measurements transmission and reflection 5 6 Sweden REFERENCES 1 M Sjoholm G Somestalean J Alnis S Andersson Engels and S Svanberg Analysis of gas di
44. ditionally different types of measurements on fruits balsa wood and volcanic rocks were made to evaluate possible applications for the technique All of these measurements were performed in transmission mode and with a signal averaging 256 recordings per sample For all measurements the area of detection of the PMT was fitted to the size of the sample Thus if the sample did not cover the diameter of the PMT 25 mm black paper was used as a mask to block out light 7 1 Standard addition As mentioned before the GASMAS technique estimates the equivalent mean path length bv using the standard addition method This could be performed 56 Chapter 7 Experimental work with or without using a sample comparing only the slopes The PMT was used as a detector for this method When the different slopes are known for different sensitivity settings on the lock in amplifier a standard addition measurement does not need to be done for each measurement 7 1 1 Sample dependency The incoming laser light to the PMT was too intense and thus saturated the PMT if an attenuator was not used A standard addition measurement was performed on polystyrene foam a folded paper and an optical filter to investigate which sample that was the most suitable attenuator Method The method to determine the standard addition curves is nothing but a length dependency test The GMS signal was measured and calculated for different distances of air for differ
45. done using a Peltier element in contact with the diode laser The Peltier element is handled with the help of a temperature controller Vary the laser temperature 10 25 C around 25 C at the same time as you observe the signal on the oscilloscope The information in the data sheet will give you a hint around what temperature you will find 794 7 nm The temperature regulator works slowly have patience and give it time to settle at the temperature you set If you are lucky you will directly sce that the rubidium cell absorbs laser light at a specific wavelength If not the absorption lines can be hidden in a mode jump see Fig 4 By changing the injection current a bit the positions of the mode jumps can be moved It can take a while even for an experienced experimentalist to find the absorption lines If you are really unlucky the entire laser might have to be exchanged If the oscilloscope has problems with triggering i e difficulty giving a stable signal you can connect the trig output of the function gen erator SYNC OUT or AUX OUT to the external trig input of the oscilloscope First control where the level is for 100 absorption on the oscilloscope screen Thereafter put a 50 grey filter OD 0 3 in front of the detector Keep the grey filter with 100 times damping in the beam path If the signal i e the ramp on the oscilloscope is damped with 50 as well the detector part functions linearly Investigate
46. e performed to study the absorption signal On both of the two diode lasers tested absorption signals were found but one had nearby mode jumps Since oxygen is homogeneously spread throughout the air a certain distance of air is equal to a certain oxygen concentration A linear relation procedure was performed to determine that the absorption signal was linearly dependent to the distance of air By measuring the GMS signal at varying path lengths a linear relation was determined for the absorption signals obtained with the two lasers tested Finally a wavelength meter was used to determine the wavelength of the laser and thus which line being responsible for the absorption of the light According to the wavelength meter the operating wavelength for the best absorption signal found was 763 31 nm at 38 08 C at an injection current of 37 4 mA suggesting that the absorption line observed was P7Q6 20 4 2 Finding suitable lasers for rubidium Rubidium is an alkali metal commonly used in absorption spectroscopy Its concentration in a sealed off cell can be investigated by observing the so called D or De lines see Fig 4 2 The search for absorption signals either the D or D line was done by letting the light from a diode laser pass a 7 cm glass cell containing rubidium gas and observe the output light with a photo detector The rubidium vapor pressure at room temperature is only 1077 mmHg but due to the very high transition pro
47. e as in e g CD players and laser printers It gives a maximal output of 10 mW which is enough to be harmful to our eyes lasers are considered harmful above 0 4 mW The wavelength 794 7 nm lies just within the limit of the infrared region and the usual sensibility of the eye is therefore very low Thus the light from the laser can seem very weak but still be so intense that it is harmful At the laboratory there are safety goggles and an IR card With the help of the IR card it is possible to localize the IR laser beam Please consider the fact that wrist watches and other reflecting parts can cause hazardous reflexes when adjusting the laser beam hots diode a eop Dinde sser i Figure 3 The diode laser is mounted within a hermetically sealed capsule The photo diode is used to monitor the laser intensity The laser is of the single mode type i e it emits light with only one fre quency and it has a line width of about 50 MHz The emission wavelength of the laser about 794 7 nm is determined primarily by the band gap in the pn transition By changing the temperature the band gap is changed and also the optical path length is changed a bit Due to the shape of the resonator the cavity and the amplification profile the laser sometimes does a jump in frequency a so called mode jump This means that the wave length is changed completely for diode lasers of GaAlAs type with about 0 3nm C Between the mode jumps the wavelength
48. e of the lines merging Thus it does not help if you have access to a narrow band tunable laser unless you can eliminate the Doppler width Even if we successfully and thoroughly eliminate the Doppler width the spec tral lines still have a certain frequency width as a consequence of Heisenberg s uncertainty relation the so called natural line width States where the atoms are for a long time the ground state and meta stable levels will become very narrow in energy The frequency width Afy of an optical transition to the ground state is determined by the lifetime 7 of the excited state For transitions to or from the ground state a spectral line will get a FWHM A fn determined by 1 Afw INT With a tunable dye laser with a frequency width of a couple of MHz it is possible to measure spectrally really sharp lines without the bandwidth of the laser broadening the recorded lines A diode laser however has a bandwidth of round 50 MHz which rarely becomes neglible in relation to the natural line width and the separation between the spectral lines Thus we have to pay attention if it is the line profile of the atoms or an instrumental broadening or a combination of them that is being observed The diode laser During the laboratory exercise we will be studying Rb with the use of a diode laser cmitting infrared light at the D line 794 7 nm A diode laser is shown in Fig 3 and is an GaAlAs diode laser of the same typ
49. e perpendicular to the laser beam Discussion Since the equivalent mean path length is shorter in the smaller slabs this suggests that the scattered light is prevented by the physical limits of the slab When the equivalent mean path length values are stabilized this suggests that the sample could be seen as an infinitely wide slab A source of error in the measurement is that different detection areas were used 60 Chapter 7 Experimental work 7 2 2 Different thickness A measurement was performed on polystyrene foam slabs with different thick ness to investigate the effect of thickness of a measured sample Method For this procedure 7 quadratic 25 x 25 cm pieces of polystyrene foam were used The sample thickness was increased by successively putting one on top of the other for each measurement The equivalent mean path length was calculated for each thickness increment Results In Fig 7 3 it is indicated how the equivalent mean path length measurements follows a second degree polynomial fit y 6 4 16 2t 5 6 0 1 2 3 4 5 6 7 Polystyrene foam thickness cm Figure 7 3 The thickness dependency of the sample for scattering in polystyrene foam The measurement samples follow the second degree poly nomial fit given in the figure Discussion The relation between the thickness of a polvstvrene foam sample and the equivalent mean path length appears to be a polynome of second degree 7 3 Drying balsa wood
50. e project the aim has been to find solutions suitable for the country of Ecuador T he survey measurements on fruits showed which fruits that were suitable for future studies Evidentially we did not test all the fruit types in Ecuador While talking with the potential collaboration partners at the Department of Food Science for example we got to hear about a root called jicama This root has interesting sugar properties that might make it possible to produce a sweetener that diabetics can use Maybe it can also be used to produce a sweetener for people who wants to diet Hopefully the GASMAS system at EPN will be a tool in understanding the maturity stages of the jicama root which is a problem at this moment The fruits we performed successful measurements on can be further and more profoundly investigated in the future We have full confidence in that the staff at EPN will find out interesting and exciting experiments to be made Hopefully the GASMAS system will help in solving the vast problem of understanding the ripening process occurring while transporting the fruits This might provide the tools for increasing the possibilities to export the fruits The gas content of volcanic rocks can give important and interesting geo logical information so the investigation of these rocks will be an interesting objective for the future new investigation needs to be done to determine the thickness needed to be able to let enough laser light through
51. e to find them in products such as CD players bar code readers and within optical communication systems 8 The wide use of diode lasers has made them mass produced and thus relatively cheap with prices ranging from 1 a piece 9 Although since they are normally produced for commercial application purposes and not for research experiments they might not have exactly the wavelength or features sought for They also have a property referred to as mode jumps see Sect 2 2 4 which limits the wavelengths available 8 There are custom made diode lasers on the market guaranteed not to have mode jumps However these Distributed Feedback lasers or DFB lasers are more expensive 10 By testing several mass produced lasers it is possible to find lasers suitable for research at a low price Another advantage of diode lasers in comparison to other laser types is their size The diode lasers including shielding and connectors are normally 2 2 The diode laser only the size of a green pea see Fig 2 3 This enables possibilities to build compact equipment Figure 2 3 A diode laser and a green pea Other attractive features of diode lasers are their energy efficiency and easy operation The major advantage of diode lasers for spectroscopic purposes however is their ability to be tuned in wavelength The ability to scan the wavelengths around an absorption line is a key feature for absorption spectroscopy 4 2 2 3 Basic principle
52. eaks in the upper part of the figure is 141 MHz Calculate the corresponding frequency separation with the use of Fig l and compare the results with each other Also determine the Half Width at Half Maximum HWHM Af measured at half the height at the highest peak of Rb Compare your result with the natural HWHM task number 5 Sketch the shape of the spectrum in Fig 2 if the peaks are Doppler broadened Use the result from task 5 Is it then possible to see some thing from the hyperfine structure of the ground state 5s S 2 and 2 the excited state 5p Py j2 respectively Estimate the HWHM of the peaks with the help of the separations in Fig 1 and the Doppler width Theory Hyperfine structure of rubidium During this laboratory exercise we are going to study the so called D line Sometimes however the Da line is studied The original laboration instruc tion which this instruction is translated from is focuscd on the D line for example Thus the theory and the examples in this instruction will be fo cused on the Da line The D line is an historical denotation for the transition between 5s 7S 2 and 5p P3 j2 The denotations D and Da do not mean anything They only refer to that the resonance transitions are appearing in pairs for all alkali atoms Think about the yellow double line of sodium at 589 6 and 589 0 nm for example In fact rubidium has two naturally existing isotopes Rb 73 which is
53. ed booking plane tickets buying fruit samples etc Experimental work includes mounting the equipment searching for and fixing errors calibrating the equipment and doing experiments Report writing includes everything having to do with producing the written report such as information searching picture making and of course the Task Marta Christoffer Preparatory work 50 50 Administrative work 50 50 Experimental work 60 40 Report writing 50 50 Technology transfer 50 50 LabVIEW programming 0 100 Miscellaneous 50 50 Table A 1 The distribution of work for the two participants of the project 88 Chapter A Work responsibilities actual writing of the report This post also includes the preparation for the presentation Information activities include making the two laboratory exercise instruc tions having presentations making a poster making a website and writing a user s manual LabVIEW programming includes the development of a computer program using LabVIEW to be used together with the equipment Miscellaneous includes everything not possible to put under the other sub jects such as meetings study visits building shelves for the electrical equip ment etc Table A 1 shows the different parts of the project and the percentage per formed for the participants As can be seen the majority of the parts are performed by both of the authors However different aspects of the parts have been done de
54. effect to sweep the wavelength of the laser with the help of a function generator The function generator can generate a current ramp of which the frequency and slope can be varied A fast current ramp gives a fast sweep in wavelength which makes it possible to observe an atomic spectrum in real time with an oscilloscope Thus the wavelength of a diode laser can be altered by varying the tempera ture of the laser capsule or by varying the injection current In the laboratory exercise temperature sweeping is used to coarse tune and the injection cur rent is used to fine tune the wavelength The temperature regulator can vary the temperature of the laser between 15 C and 60 C with the help of a Peltier element This allows for a sweepable range of about 11 nm The Fabry P rot interferometer To be able to determine the frequency separation between the spectral lines that the laser light interacts with we have to generate a frequency scale A small part of the laser light is sent through a so called Fabry P rot etalon When the frequency of the laser is changed we get a number of interference fringes equally spaced see Fig 2 The distance in frequency is called the free spectral range of the Fabry P rot interferometer In the laboratory exercise an etalon of diffraction index n and the length L is used whose both end surfaces have a highly reflecting layer The condition for constructive inter ference is 2nLcos mA which reduces to
55. en our friend and mentor providing us with tips and enthusiasm throughout the entire project Special thank to your visit to Quito where your presence gave new inspiration to a project that sometimes felt impossible Prof Edy Ayala and Dr C sar Costa both have been very helpful Their help extended far beyond their professional duties making us feel very welcome to EPN and the country of Ecuador Without Prof Ayala the equipment would still be at the Ecuatorian customs Thank you Dr C sar Costa for your friendly welcoming and your help with corrections of the report Also thanks to Christian Santa Cruz who helped us with many practical issues and questions A very special thanks goes to Yolanda Angulo who was with us every day during the project in Quito Her questions made us stay on our toes her help in the lab and with the Spanish was most welcome and her presence always cheerful Mats Andersson and Mikael Sj holm helped us many times in Sweden Mats was also very generous with his knowledge about LabVIEW and electron ics He made the task of making a LabView program possible to handle by creating a start program for us and then kept on helping us with problems occurring during the development Thanks to our Ghanaian friend Benjamin Anderson who was always very eager to help us while we where doing our preparations in Sweden We wish you good luck on the project of assembling your own GAS MAS setup in 82 Chapter 8 S
56. en absorption signal was continued for 12 more hours after making the cuts 65 68 Chapter 7 Experimental work the environment Thus the real aim of this experiment was to show the possibilities to measure on the fruit 7 6 2 Results Fig 7 8 indicates a rapid increase in oxygen absorption signal when the banana was cut Leq cm 0 100 200 300 400 500 600 Time min Figure 7 8 Measurement points showing the equivalent mean path length Leg for a banana before and after its ends were cut off 7 6 3 Discussion The oxygen absorption signal increased when the ends were cut off Even though there might have been a difference in scattering coefficient when the ends were cut off maybe due to drying it is suggested that the change in signal is due to a variation in oxygen concentration 7 7 Peeled apple Apples are pleasant samples to work with when doing GASMAS measure ments They give a nice absorption signal and are easily manipulated Stud ies on gas exchange in apples has been done before showing the possibilities 7 7 Peeled apple 69 to investigate this process using both transmission and backscattering geom etry 17 18 7 7 1 Method An apple was measured untreated for about 30 minutes then approximately 60 of the skin was peeled off while the apple remained under the laser light The reason for the latter was not to change the measurement point The apple was measured for additional 10 hours without s
57. en the measurement was finished 7 33 Discussion The increase in oxygen absorption signal suggests that the water inside the wood is replaced by air Both the molecular oxygen concentration and the scattering properties probably changed when the piece of wood was drying giving an increase to the oxygen absorption signal A source of error was probably the fact that the zero level for the normal ization was only measured before starting the measurement and not con tinuously It has been observed that dry balsa wood allows more light to penetrate than wet balsa wood 7 4 Fruit and vegetable overview Ecuador is a country full of different exotic fruits and vegetables providing income to the country through export Thus investigating oxygen concen tration in fruits is of great importance The availability of oxygen affects the ripening process and the quality of the fruit since the gas is a part of the metabolic processes of respiration The respiration rate of the fruit is proportional to the organic breakdown So if the oxygen concentration inside the fruit is decreased the life time of the fruit is prolonged However if the oxygen concentration decreases below a critical level fermentation starts which initiates a rapid decay of the fruit Thus optimizing the concentration of oxygen is of great interest during the postharvest time of the fruit 17 An investigation was done on several types of fruit and vegetables to test if GASM
58. ence and conduction bands The left figure shows the bands without a bias voltage and the right shows the bands with a bias voltage being applied over the laser The homojunction lasers have a major drawback they cannot work at room temperature This is due to the large thickness of the active medium and losses from absorption in the junction This results in the need of a very high current at room temperature for the diode to be able to lase 7 Heterojunction lasers The problem the homojunction lasers have with the operation temperature is solved for heterojunction lasers T hey have an active layer also a semicon ductor sandwiched between the two semiconductor layers with higher band gap energies see Fig 2 5 Since the photons created in the active layer do not have the energy corresponding to the surrounding band gaps the photons will not be absorbed This allows the laser to operate at room temperature 7 Figure 2 5 The band gap structure of a heterojunction laser showing the active layer in the middle with the energy gap Eg The surrounding band gaps E2 have a higher energy gap than the active layer 2 2 The diode laser Laser production First the light will be spontaneously emitted and amplified with the help of a gain medium in an optical resonator a so called cavity The cavity is formed by having a reflectance of about 30 on the cleaved faces of the semiconducting material The spontaneously emitted photo
59. energy is conserved in the processes The Raman scattering is inelastic meaning the energy is changed in the process and thus the wavelength is shifted 3 4 13 14 Chapter 2 Theory 2 2 The diode laser Since the introduction of diode lasers they have become a common compo nent in our daily life Their size price and ability to easily tune in wavelength have also made them common in absorption spectroscopy Their semicon ductor structure allows them to lase at low power and room temperature However the diode lasers also have product specific disadvantages like beam divergence and mode jumps 2 2 1 History The first version of the diode laser was developed simultaneously in 1962 by four different and independent science groups The first diode lasers devel oped were homojunction based units and had to operate at temperatures of only a few Kelvin Later during the same decade the heterojunction based lasers were discovered and made more or less the homojunction based lasers obsolete The heterojunction lasers function at room temperature which opened possibilities for many new applications 7 During the 1980 s this laser type experienced a rapid increase in commercial use particularly re garding telecommunications 8 One application among many that became more available because of this was laser spectroscopy 4 2 2 2 Advantages Diode lasers are the most commonly used lasers in the world today It is possibl
60. ent types of samples attenuating the signal Results Table 7 1 shows the different equivalent mean path lengths and the slopes of the standard addition curves obtained for the different samples Sample Slope a u cm Le cm Polystyrene foam 0 52 30 2 Folded paper 0 68 2 05 Optical filter 0 71 0 57 Table 7 1 The different standard addition slopes obtained and the calculated equivalent mean path length for three different light attenuating samples Discussion Standard addition gave as expected different equivalent mean path lengths depending on which sample being used The obtained slopes also varied but should theoretically be the same Particularly the slope obtained when 7 1 Standard addition using polystyrene foam deviates and this is probably due to the relatively small difference measured upon a large signal and the limit in resolution on the oscilloscope The folded paper which is very thin gives a relatively large equivalent mean path length and this can maybe be explained with the possibly high scattering in the paper and the reflections between the sheets The equivalent mean path length obtained when using the optical filter can be seen as the internal offset of the system since the filter provides no additional offset 7 1 2 Lock in amplifier sensitivity setting dependency The slope of the standard addition curve depends on the sensitivity setting of the lock in amplifier This mathematical relation was investi
61. er investigation some different types of volcanic rocks were investigated 7 9 1 Method The different types of rocks see Fig 7 11 were provided by the Department of Geophysics at EPN They also sliced the samples so that the samples would be geometrically suitable for measurement with GASMAS The rocks were measured to see if a signal could pass through the rock and the quality of this signal was determined 7 9 2 Results Table 7 3 shows that the GASMAS system was not able to measure an oxy gen absorption signal in the volcanic rock samples provided None of the V1 Chapter 8 Summary and conclusions There have been four distinct parts involved in this project preparations technology transfer assembly and experiments 8 1 Results of the project 8 1 1 Preparations The preparations took place in Sweden and resulted in the finding of different diode lasers suitable for observing absorption lines for oxygen and rubidium The diode lasers were later used in Ecuador but different absorption lines were observed due to changes in laser performance The preparations also allowed for a successful delivery of the equipment from Sweden to Ecuador with the help of ISP in Sweden and Prof Ayala in Ecuador 8 1 2 Technology transfer While waiting for the equipment to arrive and during the entire time spent at EPN efforts were made to transfer the technology and know how about the equipment This was done by holding a presentation
62. erial An investigation was performed show ing the influence of physical width of the sample Another measurement performed was to see the relation between thickness of a sample and the equivalent mean path length The results showed an effect of limited width of the sample that became negligible after a certain dimension suggesting that the slab could be seen as infinitively wide The results obtained be tween the thickness of the slab and the equivalent mean path length followed a second degree polynomial fit However according to earlier investigations this relation should be quadratic 12 8 2 Future work Balsa wood A measurement of drying balsa wood was performed to investigate the sup posed increase of oxygen in the wood as it was drying The result showed an increase in oxygen absorption signal suggesting that when water vaporized from the wood oxygen entered Fruits First measurements were made on several different types of fruits to provide an overview of which fruits that gave good enough signals for performing measurements Papaya banana tree tomato and apple were further inves tigated Measurements over time were made when the skin of the fruits was partially removed All fruits showed a change of oxygen inside the fruit over time Apple and banana increased their oxygen absorption signal while the opposite was observed for the papaya The reason for this difference might be stemming from an initially different
63. g the EPN for about 10 days supporting the project The objective of this presentation was to give a detailed and lively introduction to the system The intention was to present the possibilities of the system to specially invited researchers at the university in order to start a discussion on possible future collaborations Appendix C shows an Abstract of this presentation 5 3 Laboratory exercises 5 3 Laboratory exercises Involving many students in the GASMAS activities including getting hands on experience with the equipment was another important objective By having laboratory exercises the system will be used by many people and hopefully some students might get interested in continuing to work with the system at the summer practice Diploma and Master Thesis level A laboratory exercise made and frequently used in Sweden for rubidium ab sorption spectroscopy was translated from Swedish into English and further modified see Appendix D A laboratory exercise for understanding the ba sics of GASMAS was prepared and carried out together with students at EPN see Appendix E 5 4 Manual The experiences from the establishment of the laboratory were discussed and collected in a user manual see Appendix F This manual is meant to facilitate the technique and the set up to future users It includes a brief theoretical introduction and a detailed description of the different parts of the set up at EPN and its operation Also some basic s
64. gated Method Different standard addition curves were measured for different sensitivity settings on the lock in amplifier The optical filter was used to attenuate the light to the PMT Results Fig 7 1 shows the standard addition curves obtained for different sensitivi ties The results suggest the following dependency between the slopes of the standard addition curves and the corresponding sensitivity S2 ksi oy ksa A where kg is the slope obtained with the sensitivity setting S1 and kgs is the slope obtained with the sensitivity setting 52 Discussion The mathematical relation found between the different sensitivity settings used does not correspond perfectly to the data but well enough to believe that the mathematical relation is correct The small discrepancy is thought to be the result of inherent inaccuracies in the measurements of 08 Chapter 7 Experimental work sensitivity 10 y 1 4498x 0 86344 sensitivity 20 y 0 71943x 0 4131 an sensitivity 50 a y 0 32269x 0 18862 GMS a u sensitivity 100 y 0 14512x 0 085225 sensitivity 500 y 0 03307 1x 0 018341 sensitivity 1000 y 0 01622x 0 010109 Licm Figure 7 1 Standard addition curves for different sensitivity settings on the lock in amplifier The left picture shows all the standard addition curves and the right one shows a detailed part of the
65. gnals The lock in signals correspond to the second derivative of the direct signal signal containing two AC signals one with the added frequencies wg wr and one with the subtracted frequencies wa wz Vesp VeigStn Wpt T PR Vi sin wyt F pL 1 5 Vsig Vic08 wr wr t Pr PL 1 5 VsigVLcos WR EN Wr Ji EN Pr EN PL The multiplied signal Vpsp is filtered with a low pass filter leading to the AC signal with the added frequencies being eliminated The remaining signal will be a DC signal since wr wz 1 Vesp 5 Vsig VLCOS wR wL t PR YL 1 VsigVLCos Pr Yr Vesp X Vsigcos pr PL 2 5 GASMAS This DC signal V cos r yz can be adjusted to its maximum amplitude by changing the phase to yr yy The DC signal Vpsp is proportional to the sought signal and denoted in this report as the lock in signal There are three important properties of the lock in amplifier when optimizing the signals the phase the sensitivity setting and the time constant The phase setting changes yy and thus also the amplitude of the obtained lock in signal The sensitivity setting changes the amplification of the pre amplifier The time constant refers to the time constant of the slope of the low pass filter 2 5 GASMAS GASMAS GAs in Scattering Media Absorption Spectroscopy has common features with other types of gas absorption spectroscopy It uses the three ba sic modules the light sour
66. h as polymers ceramics wood our body and fruits using narrow banded radiation from tunable diode lasers The possibility to measure the gas contained in such materials relies on the fact that solid materials and liquids have broad absorption features with linewidths normally not sharper than 10 nm while free gases typically have a linewidth 10000 times sharper In the porous materials the radiation is heavily scattered which results in a not well defined path length for the light that passes through The mean path lengths for the light are frequently orders of magnitude longer than the geometrical dimensions of the sample The setup is based upon diode laser spectroscopy with three main components a light source sample and detector Molecular oxygen has so far been studied by using a diode laser at 760nm The possibility to use the technique for postharvest monitoring in the fruit industry has been studied by the technique presented Furthermore for medicine applications human trials on the maxillary and frontal sinuses have been preformed for possible use in sinusitis diagnostics see Fig 1 By changing the diode laser it is possible to study other gases such as CO and H20 The equipment will very soon be available in Quito for research and education purposes 1 5 Oxygen content a u Ol 03 Off On Fig 1 Oxygen concentration measurements of the frontal sinuses Appendix D Rubidium laboratory exercise Laboratory exerc
67. harmonic is studied The two signals from the generators are added This is done by a power splitter Mini circuits ZFRSC 2050 Depending on how the device is used it can either take one signal and make two exact copies of it or add two signals into one However using the device in this last manner the amplitudes of the input signals are halved 3 4 Detectors used a solid state photo detector and a photomultiplier tube PMT both creating a current proportional to the detected light The PMT is a more sensitive detector and has a larger detection area which is preferable when absorption in scattering materials is studied Each detector could be used for both set ups but normally the PMT is preferred for the GASMAS set up and the photo detector is preferred for the rubidium set up 34 Chapter 3 Equipment 3 4 1 Photo detector A battery driven photo detector Thorlabs DET110 is used to detect the laser light in the rubidium set up and for the process of selecting usable lasers The photo detector uses a photodiode to create a current proportional to the detected light The detector can be saturated by a too high intensity of light 3 4 2 Photomultiplier tube In the GASMAS set up the intensity of the detected light is normally very weak since it is attenuated in a scattering medium In order to detect this small intensity of light a photomultiplier tube Hamamatsu R5070A is used since it is an extremely sensitive detector T
68. he absorption signal and plots the values as well as saves them into a file for further manual investigations The functions of the program opened up new possibilities for automated and unattended long time measurements Measurements that was not possible to perform earlier could now routinely be accomplished There were three major functions of the program called Standard addition Single step and Continuous measurement The Standard addition function was made for discrete measurements where the conditions were changed be tween each measurement It calculates a GMS value from a signal by making a zero level measurement for each new measurement A zero level measure ment is defined by us as the measurement of background light It is measured when the laser is turned off The Standard addition function is preferred for measurements that need a high accuracy such as standard addition The Single step function was also made for discrete measurements and works very similarly to the Standard addition function The difference is that it does not make a zero level measurement for each new measurement Hence this func tion is faster but less accurate than the Standard addition function The last function and mavbe the most important one is Continuous measurement This function is what allows the user to make automated and unattended long time measurements The user measures the zero level through the pro gram enters the elapsed time of the procedure
69. he large detection area is also an advantage since the light is scattered A photomultiplier tube consists of a photocathode an electron collection system an electron multiplier section dynodes in a cascade manner and an anode When the incoming light hits the photocathode electrons are emitted If a high voltage is put over the PMT the electrons get directed and accelerated towards the dynodes and the anode When striking a dynode the electrons produce secondary electrons creating a cascade The anode will receive the electrons and create a current 27 The high voltage is provided from an external high voltage supply in the GASMAS set up It is of high importance not to let the signal current from the PMT reach too high values since this can destroy the device For the setup the output current was measured over a resistance and displayed with an oscilloscope The maximum output current from the PMT is generally 100 yA and it is usually linear up to 10 yA With a too high current the PMT gets destroyed because of ohmic heating 3 5 Lock in amplifier The lock in amplifier used in the GASMAS set up Stanford Research Sys tems SR810 is digital It has a reference channel frequency that ranges between 1 mHz and 102 kHz 26 The properties phase sensitivity and time constant see Sect 2 4 3 are frequently used to optimize the properties of the lock in signal 3 6 Oscilloscope 3 6 Oscilloscope In order to observe the signals
70. hen you have set the frequency and amplitude you connect the generator signal to the diode laser driver modulation input MOD IN CAUTION DIODE LASERS ARE EXTREMELY SENSITIVE FOR TRANSIENTS i c clectrostatic discharges cable reflexes ctc Never shut off the current to the function generator or disconnect it from the laser driver unit as long as the diode laser is turned on Make it a habit to always turn off the injection current the button at the upper right not the main current button at the lower left to the diode laser before you make a change in the experimental set up 10 Put on the safety goggles Adjust the injection current to the recom mended value Jop The laser is now ready to be used Focus the laser beam with a lens onto the detector a photo diode Connect the detector via the current voltage amplifier to the oscillo scope and investigate the detector signal You will see how the laser intensity varies with the injection current to the diode laser It is possi ble that the light intensity becomes so great that the detector becomes saturated saturation will happen if you really focus the laser beam onto the detector Place a grey filter attenuating the laser intensity 100 times OD 2 0 in the beam path Insert the rubidium cell in the beam path Be careful with reflexes The wavelength of the diode laser can now be coarsely tuned by altering the temperature of the entire diode laser capsule This is
71. illa taxo granadilla and aloe vera all were possible to test 63 64 Chapter 7 Experimental work Fruit Leg cm Thickness cm Signal appearance Papaya 10 13 Descent Banana 22 3 9 Good Tree tomato 1 4 5 Good Potato 1 0 5 Unsatisfying Hawaiian papaya 7 No signal Coconut 10 No signal Avocado 6 No signal Passion fruit 6 1 5 0 Good Physalis 0 9 2 Descent Horito 22 20 Descent Cassava 6 No signal Pitaya 1 1 4 5 Unsatisfying Apple 20 6 5 Good Guayaba 9 9 9 9 Unsatisfying Naranjilla 1 5 9 9 Descent Taxo 3 5 3o Descent Granadilla 23 6 5 Descent Aloe vera 1 0 2 Good Table 7 2 The fruits tested and their corresponding equivalent mean path length Leq the thickness of the sample at the point of measurement to the point of detection and the appearance of the signal in relation to the other fruits no signal unsatisfying descent and good The denotations of the appearance of the signal are the subjective estimate of the authors and is related to the signal to noise ratio 7 5 Cutting a papaya further However some of these fruits only had a descent signal resulting in a larger uncertainty in the GMS value The GASMAS technique was not at all able to determine the oxygen concentration in coconut avocado and cassava since the laser light would not pass these samples This is probably due to the dense skin of the coconut and cassava and the large and dense kernel of the avocado It is possible to use the GASMAS technique to in
72. injection current for different temperatures Un cion Figure 9 The divergence features of a diode laser Another more important disadvantage of diode lasers is that they tend to mode jump i e making discrete jumps in wavelength see figure 10 As with any laser the output wavelength is the one corresponding to the standing wave in the laser cavity When the laser is tuned in wavelength the gain curve is shifted resulting in mode jumps To fully use the laser properties it is always important to only have one wavelength in the output from the diode laser a so called single mode laser Multi mode operation occurs when more than one standing wave in the cavity can lase equally well A A TEN Mode jump LT Figure 10 Diagram showing mode jumps making discrete jumps in wave length Absorption Atoms and molecules absorb certain characteristic frequencies or wavelengths This is an effect of their electronic shell structure and the vibrational and rotational energy levels If the photon energy i e if the frequency or the wavelength of the light is matching the energy separation of the atom or molecule it may use the energy and get excited and the photon is absorbed Every atom and molecule has a unique set of absorption lines i e its fin gerprint Absorption can theoretically be described by the Beer Lambert law It states that the intensity of the incident light J attenuates exponentially as the light tra
73. ions used for normal ization are given in the figure 2 5 3 Standard addition In GASMAS the standard addition method is introduced to determine an equivalent mean path length The method is well known in physical chem istry and it relates the absorption signal to that of absorption in free air The basis of this calibration technique is that a linear relationship between the absorption signal and the oxygen concentration is expected since the absorption signal is only a few percent of the signal An equivalent mean path length Leg can be extrapolated by letting the laser light travel a known distance through a well characterized oxygen rich medium such as normal air and determining the increase in absorption sig nal see Fig 2 15 For a scattering sample this would correspond to the distance the light would have to travel through air to obtain the same signal For this reason the equivalent mean path length can be longer or shorter than the thickness of the measured sample 1 The reason for using the standard addition method and the equivalent mean path length unit is to transfer the rather abstract oxygen absorption into a tangible unit The equivalent mean path length depends on both the concentration of molecular oxygen and the scattering coefficient of the sample In a highly scattering sample the light will travel a longer distance along more compli 2 5 GASMAS GMS a u a Las L cm Figure 2 15 Several standard add
74. is changed 0 06 nm C or in another words 30 GHz C In Fig 4 it is shown how the wavelength depends on the temperature to AA a Ruticium absorption 380 2 rr RIIIE PARMA RANIA ORI EE ERT TE RTT TE R a A A en CONurous Wavelength shift a ae ier Une aoe ee T t es aie Figure 4 The emission wavelength of the diode laser 1s dependent of its temperature Note the sudden changes in wavelength The light intensity of a diode laser depends on the injection current Fig 5 As long as the injection current is small the component functions as a regular light emitting diode LED When the injection current exceeds a certain threshold value denoted Jn light amplification occurs through stimulated emission i e laser action The injection current needed to generate this k effect varies a bit between lasers The recommended injection current Lop is together with other types of data given in a data sheet for each diode laser The laser in this lab can normally be used up to 10 mW but the life time of the laser will be shortened at this power and if this value is exceeded the laser might be destroyed Figure 5 The output power of the diode laser Py as a function of the injection current Ip for different temperatures The emission wavelength of the laser can also be changed by varying the injection current which is due to the fact that the temperature is changed locally in the pn junction We will use this
75. ise Diode laser absorption spectroscopy in atomic rubidium Originally written at the Department of Physics Lund Institute of Technology Sweden Translated from Swedish and modified by Christoffer Bjorkwall and Marta Cassel Engquist September 2005 Introduction The purpose of this laboratory exercise is to give a basic knowledge about diode laser spectroscopy With the use of laser techniques it is possible to study very detailed structures of the energy levels of atoms Hyperfine struc ture isotopic shifts and Zeeman splittings are examples of detailed structures that can be studied However in this lab we are going to focus on the exper imental techniques Experimental skills of course also include evaluating the results from the measurements In this laboratory exercise this specifically means understanding the definition of line width for the subject studied and for the light source and its effect on resolution and measurements of small energy differences Regarding the technique of measuring the goal of this laboratory exercise is to show how to determine gas concentrations by using lasers Laser tech niques can be used to determine everything from the oxygen concentration in a respirator to the air pollutions at a certain level above ground Dur Copyright for the text and all the picture belongs to the Department of Physics Lund Institute of Technology Sweden ing this laboratory exercise we will investigate the
76. ition measurements giving the equivalent mean path length through an extrapolated line cated pathways Hence there will be a larger probability to find molecular oxygen The real concentration of molecular oxygen in the sample Csm can then be related to the concentration in free air Cair by using Leg according to Cupless n Cand ans 2 9 where Lsm is the actual optical path length traveled by the light inside the sample 12 2 5 4 Restrictions A very important limitation one has to deal with when performing gas ab sorption measurements in the general case is the absorption of other com pounds The interferences depend on the sample type and the spectral re gion being analyzed Among the most important interferences is the ubiq uitous water vapor Water is widely distributed in all types of biological tissue and strongly absorbs light for wavelengths larger than 1400 nm 1 Hemoglobin absorption practically eliminates all light transmission for wave lengths shorter than 600 nm The oxygen absorption lines that GASMAS uses are between these wavelengths around 760 nm making it possible for the light to penetrate human tissue without substantial absorption by these compounds see Fig 2 16 The wavelength range between 600 nm and 1400 nm is called the tissue optical window 29 30 Chapter 2 Theory Log Absorption coefficient 300 500 700 1000 2000 3000 Wavelength nm Figure 2 16 The absorption in human tissue due t
77. ived it was assem bled and optimized as planned During this part a local Diploma student Ms Yolanda Angulo was taught about the systems and was step by step involved in the mounting and investigation processes Together with our local supervisors at EPN Prof Edy Ayala and Dr C sar Costa different investigations were planned and performed Measurements were made on fruits balsa wood volcanic rocks and polystyrene foam with promising results Additionally an introductory laboratory exercise was created for GASMAS and applied A laboratory exercise for the rubidium set up was also trans lated from a corresponding activity used at LTH A program was developed in LabVIEW making it possible to perform automated and unattended long time measurements As a help for future researchers on GASMAS at EPN a user manual was made In order to make information about the systems and laboratory exercise instructions easily accessible a website was constructed and uploaded on the EPN website 1 5 Outline This report contains a theory chapter Ch 2 that treats the basic component of diode laser absorption spectroscopy interaction between light and matter 10 Chapter 1 Introduction diode lasers and modulation techniques The theory chapter ends with a de scription of the GASMAS system and method Ch 3 describes the different parts of the equipment The other chapters are placed in chronological order as they followed in the project
78. kin 7 7 2 Results 0 100 200 300 400 500 600 700 Time min Figure 7 9 The measurement points and an exponential fitted curve indicat ing the increase in oxygen absorption signal over time for a peeled apple As can be seen in Fig 7 9 the oxygen concentration increased when the apple was peeled The measurement points are fitted with an exponential curve and the time constant was estimated to 80 min 7 7 3 Discussion There was a significant increase in GMS signal after the apple was peeled This suggests that oxygen entered the apple when the skin was peeled off Chapter 7 Experimental work Thus the skin functions as a membrane keeping a certain oxygen concen tration in the body of the apple Similar results have earlier been observed 17 18 7 8 Nitrogen exposed tree tomato Tree tomato is a fruit very similar to the common tomato but slightly different in size and shape It is also much sweeter and is therefore common in fruit juices in Ecuador The technique to study gas exchange in fruit by pre treating the fruit by immersion in nitrogen gas has earlier been performed on apples 17 but never on tree tomatoes 7 8 1 Method A ripe tree tomato was put in a glass tank of approximately 25 liter filled with nitrogen at atmospheric pressure After about 5 hours the tomato was removed from the tank and instantly the oxygen absorption signal in the fruit was started to be measured The objective was to observe
79. l levels follow the rule AF 0 1 corresponding transitions at Rb two groups with close lying peaks are visible With rubidium the separation in frequency between the spectral lines are at most 7 GHz 0 014 nm It can be seen in both the experimental recording Fig 2 and in the energy level diagram Fig 1 Practically it is not very easy to detect the small differences in wavelengths caused by the hyperfine structure To better understand the problem we will briefly discuss line widths Line widths During the laboratory exercise we are using a cell containing a dilute rubid ium gas The laser light getting absorbed by the rubidium gas will be more or less Doppler broadened Full Width at Half Maximum FWHM A fp for a Doppler broadened spectral line is determined by the temperature of the gas T according to T Afo Co fp 77 C 7 16 x 10 kg K7 kmo A JE E eb E so Ss A an i E be ima Y A e 2 xc 33 r E G Swe prea ae T cS a ue y E Y N PE a nie T if as Figure 2 A laser recording of the Da line of rubidium Rb and Rb made with a collimated atom beam In the figure there are also interference fringes from a Fabry P rot etalon with a free spectral range of 141 MHz Crt where M is the molecular weight and fo the frequency of the transition Since cach linc has an individual Doppler profile small structures are often hidden in our case the hyperfine structure becaus
80. large modulation amplitude gives a wider absorption signal and vice versa This is due to fact that the modulated signal observes a wider span It is also shown in the figure that an increased frequency gives rise to a smaller amplitude of the signal This might be caused by the electronic components slowing down the transfer function or the modulation frequency being too fast for the diode laser to follow The optimal settings of the modulation signal were found to be around 10 kHz in frequency and 60 mV in amplitude With frequencies under 10 kHz the signal rapidly got affected by the high pass filter with a cut off frequency of 1 6 kHz and noise 6 2 2 Situation specific noise All unwanted signals in a system that can be detected at the output are referred to as noise Included are external environmental noise and noise due to certain characteristics of the equipment Due to the situation of the laboratory at EPN and the specific properties of GASMAS some particular types of noise were identified Some of these noise sources were specific for the environment at this very laboratory As mentioned before noise types like interference fringes interference of the ramp and the noise from the lock in amplifier were detected and treated but other types of noise were also discovered 49 30 Chapter 6 Assembly Modulation frequency 10 kHz 20 kHz 30 kHz 40 kHz 50 kHz 20 mv 40 mV Modulation amplitude 60 mV 80 mV 1
81. left curve 7 2 Polystyrene foam Polystyrene foam is an easy material to work with when doing oxygen ab sorption measurements The material has high scattering properties and is full of air pockets both properties give a strong oxygen absorption signal Thus the material is useful to investigate the nature of light propagation such as scattering 7 2 1 Different width A measurement on different polystyrene foam slabs with varying widths was performed This was done to investigate how far the scattering reaches per pendicularly to the incoming beam for a certain thickness of the slab Method To investigate the scattering perpendicular to the laser beam the oxygen ab sorption signal was measured for quadratic slabs of 11 mm thick polystyrene foam with different widths ranging from 0 5 cm to 10 cm in width The 7 2 Polystyrene foam 59 equivalent mean path length was measured and calculated for every piece The detector was masked for samples smaller than the detector area Results As can be seen in Fig 7 2 the equivalent mean path length increases until the piece with a 3 cm width Thereafter the equivalent mean path length is stabilized at around 17 cm 0 1 2 3 4 5 6 7 8 9 Width of polystyrene square cm Figure 7 2 The width dependency of the sample for scattering in polystyrene foam The results indicates that the scattering in a 11 mm thick polystyrene foam sample stretches to approximately 1 5 cm to each sid
82. length Fig 5 shows a diode laser together with a match pea and a peanut Figure 5 The size of a diode laser in comparison with a match a green pea and a peanut Diode lasers or semiconductor lasers are produced as a compound of differ ent materials The materials used depend on which wavelength the laser is manufactured to produce It is today possible to reach wavelengths between 0 4 to 29 um with different types of diode lasers The majority of the diode lasers are made of doped materials from group III e g Al Ga In and group V e g N P As Sb in the periodic system The first types of lasers pro duced were the homojunction lasers but they are obsolete today in favor to heterojunction lasers Homojunction lasers have a more simple construction 4 than heterojunction laser and will be described as a mean to understand the function of a diode laser Homojunction lasers A homojunction diode laser is created by joining semiconducting materials One of the materials is n doped has an excess of electrons and the other one is p doped which means it has an excess of positive carriers called holes When a voltage is applied over the semiconducting material the electrons from the conduction band and holes from the valence band will diffuse through the interface and be able to recombine see figure 6 As a result photons will be emitted with the energy corresponding to the band gap Depletion a a lt 2 a ee Conduction
83. malize the signals in order to be able to compare results The measured 2 intensity can vary depending on the sample geometry the alignment and some other things In GASMAS normalization is performed by taking the height of the lock in signal divided by the intensity of the direct signal at the location of the absorption signature see figure 17 The normalized signal is referred to as the GASMAS signal and is denoted GMS Figure 17 Diagram showing the normalization measurement points 514 8 Sof oa 2 GMs 2 3 aie Standard addition There are many ways of estimating gas content GASMAS uses standard addition to determine an equivalent mean path length which relates the measured signal to that of absorption in free air In the standard addition procedure the absorption signal is measured for different distances between the sample and the fiber With this procedure a calibration line can be drawn and the equivalent mean path length Leq can be extrapolated see Fig 18 The equivalent mean path length defines how much a certain oxygen absorption corresponds to the distance the light would have to travel through air to obtain the same signal For this reason the equivalent mean path length 13 can actually be longer or shorter than the real thickness of the measured sample 5 d U O lt E L cm Figure 18 Several standard addition measurements giving the equivalent mean path length through an extr
84. n is a key component in the respiration process of biological tissue such as apples A high concentration increases the it very important to be able to control the molecular oxygen concentration inside the biological tissue 4 5 2 The fundamental components of the GASMAS setup and other methods for gas absorption spectroscopy are a light source a sample and a detector and oxygen consumption By measunng the oxygen concentration as a function of time conclusion can be made on gas transport These results can be helpful to invent a new way of packaging food to last longer 4 5 S Oxygen concentration a u Medical applications on sinuses Very recently measurements have been performed on the frontal and maxillary sinuses by using reflection detection Normally to determine if a person has a sinusoidal infection a r doctor needs to choose either to amp n perform an expensive CT scan or Ref Amplifier just prescribe antibiotics The results suggest that by measuring the a amount of a mdmecily by measuring oxygen diagnosing the iness can be made cheaper 6 a 9 kHz Oscilloscope Diode faser the fight source The diode laser used in the setup depends on the gas of interest In the current setup a diode laser at 760 nm is used for molecular oxygen studies The gas content can be estimated by wavelength tuning the laser across the extremely sharp absorption ime of the free gas Scattering me
85. nd the Fabry P rot fringes to a floppy disc This will later be used for an analysis of the spectra To be able to calculate the ab sorption the zero level must be present in the figure you will obtain this zero level by blocking the laser beam 100 absorption 11 Evaluation of the spectrum a Calculate the linewidths A fabs b Calculate the distances between the absorption lines to get the level splittings c Calculate the absorption in percent of the light intensity of the two Rb lines and determine the linear absorption coefficient u if we know that I x Ipe where x is the path length through the gas and Jp is the intensity before the gas cell d Calculate the atom density in the ccll Use the following approximate relation Te SIT nar abe Gy 2 Ee where Umar is the value of u at the peak of the absorption profile and T is the lifetime for the excited state The statistical weights can be These values were used for the Swedish laboratory exercise They might not apply for this laboratory exercise 12 calculated with the results G4 G 5 G4 G 3 at the strong and the weak absorption signal respectively 12 Think about your results Do your values make sense Compare with the results of earlier laboratory exercises ask your laboratory super visor for these results and the values you calculated in the prepara tory questions atom density Doppler width line width and distance Comment explain
86. nm mA 11 Practically this means that temperature is used for coarse tuning and the injection current is used for fine tuning of the wavelength 2 3 Absorption spectroscopy However there are problems involved in tuning diode lasers The standing wave in the laser cavity determines the wavelength sent out from the diode laser In order to have a well defined wavelength the diode laser has to operate in a single mode Single mode operation means that there is only one standing wave in the cavity and thus only one wavelength being emitted The opposite condition multi mode operation occurs when there are more than one standing wave simultaneously in the cavity This is due to the frequency separation between the modes being smaller than the width of the gain profile of the laser 7 A multi mode behavior results in more than one wavelength competing in the output 2 3 Absorption spectroscopy With absorption spectroscopy it is possible to investigate a sample quanti tatively and qualitatively Every material has its own fingerprint created by the energy levels in the atom or the molecule From the results of absorp tion spectroscopy the concentration temperature and pressure of a gaseous sample can be quantified 4 An absorption spectroscopy set up consists of three main parts a light source an absorbing sample and a detector The light is sent through the sample and the output light is detected and measured as shown in Fig 2 8
87. ns being mirrored will cause stimulated emission emitting photons with the same phase and wavelength and if population inversion is provided the laser will start to lase 7 The output power of the diode laser rapidly escalates once a threshold cur rent In has been reached see Fig 2 6 Exceeding a certain level of output power will cause the laser beam to irreversibly destroy the semiconductor facets and thus the entire laser 7 Intensity tht tha Ing Current Figure 2 6 The output power as a function of the injection current for dif ferent temperatures T The diode laser starts to lase at a threshold current Lip 2 2 4 Optical properties There are some disadvantages with diode lasers for example the output beams are astigmatic assymetric and divergent The astigmatism is a re sult of the fact that the refractive index has a directional dependence The assymetrical and divergent properties are due to the assymetrical shape of the diode laser normally rectangular 1 um x 3 um in the active layer This results in a 30 40 x 10 20 divergence The beam however resembles a Gaussian profile minimizing the problem The problem with divergence of the laser beam can rather easily be handled with a collimating lens 8 17 18 Chapter 2 Theory 2 2 5 Mode jumps The major drawback of diode lasers is that they tend to mode jump These discrete jumps in wavelengths see Fig 2 7 are due to a shift in gain curve
88. nterference fringes appearing as a wavelike feature on the lock in signal Initially the laser beam was focused into the fiber with a lens but to pre 4T 48 Chapter 6 Assembly vent reflections the lens was removed This action caused a decrease in the transmission intensity from 75 to 50 60 but removed the fringes sub stanstially together with an offset caused by the air distance It was discovered that the ramp signal interfered with the modulation signal in the lock in amplifier creating a low frequency interference In order to remove this signal a pre made high pass filter with a cut off frequency of 1 6 kHz was put just before the input of the lock in amplifier When the high pass filter was introduced in the circuit the direct signal almost disappeared This can be explained by the impedance of the high pass filter 10 kQ Since the oscilloscope has a fixed input impedance of 1 MQ a voltage division occurred leaving the oscilloscope with approximately 1 of the signal By making two other high pass filters with the same frequency response but other impedances 100 kQ and 1 MQ respectively the fraction of the signal to the oscilloscope could be improved 6 2 1 Optimizing the parameters The settings of the electronic equipment in general and the function genera tors in particular are important for a well functioning system The optimal amplitude and frequency settings for the function generators were systemati
89. ny times more narrow absorption lines typically 10 times than liquids or solids 6 Thus the narrow absorption gas line occurs in a background of a broad absorption feature that comes from the scattering material Hence it is possible to assume that these background properties are constant over the range the wavelength is tuned 12 Due to the scattering the path length of the photons is not the same as the thickness of the sample This results in a more difficult approach to the Beer Lambert law There are ways of estimating the real distance the light has traveled For this purpose spatially resolved time resolved or frequency domain methods have been introduced Then by analyzing the scattering and absorption properties and using the Beer Lambert law the distance traveled by the average photon can be calculated 1 In some applications it is not necessary to know this In these cases it is enough to introduce a unit called equivalent mean path length Leg as a relative measure of the concentration of gas inside the sample see Sect 2 5 3 This unit however also depends on the scattering properties of the sample 1 For the modulation techniques in GASMAS it is convenient to use the second derivative of the signal due to the fact that the second derivative is not sensitive to the general slope of the direct signal The GASMAS technique uses a modulation frequency in the kilohertz range so called wavelength 2 5 GASMAS
90. o hemoglobin HbOz melanin and water The tissue optical window ranges between approximately 600 nm and 1400 nm where light can penetrate human tissue without being substantially absorbed by the HbO or water The arrow indicates the location of the wavelength for the A band of molecular oxygen 21 Chapter 3 Equipment There were two different absorption spectroscopy set ups sent to Ecuador one studying rubidium gas and one using the GASMAS technique The set ups consist of mechanical optical and electronical parts Many of these parts are shared between the set ups This results in that both set ups cannot be used at the same time The rubidium set up is a good and simple example of absorption spectroscopy and is therefore often used in laboratory exercises in absorption spectroscopy The essential parts are a diode laser a glass cell containing rubidium gas and a photo detector The GASMAS system see Fig 2 12 is more complex and uses modulation techniques hence it also needs a modulation generator and a lock in amplifier Since it handles smaller signals it also needs a photomultiplier tube as a detector 3 1 Diode lasers In the GASMAS set up a 5 mW commercial diode laser from Roithner LaserTechnik RLT7605MG is used These diode lasers lase nominally at around 760 nm They are commercially used for laser printers 22 and thus the supply is good and the diode lasers are relatively cheap The diode lasers used in the GASMAS
91. o referred to as derivative spectroscopy since the modulation signal gets the form of a derivative of a certain order if the modulation is small in amplitude 15 16 In Fig 2 11 the direct signal is shown for an absorption signal and a mode jump together with their resulting lock in signal if the second harmonic output is studied The lock in signals are then proportional to the second derivative of the direct signals 16 2 4 3 Lock in amplifier The key instrument in modulation techniques is the lock in amplifier This instrument uses a technique called phase sensitive detection to detect AC signals as small as nanovolt with a very good signal to noise ratio The lock in amplifier uses an external reference frequency to modulate the experimental system wp and to create an internal signal wz thus wr wp The lock in amplifier creates an internal signal Vrsin wrt yz from the input reference signal from the modulation generator The input signal to the lock in amplifier obtained from the modulated experimental system Vyigsin weat pr is amplified with a pre amplifier on the lock in ampli fier The two signals are multiplied by the lock in amplifier leading to a 23 24 Chapter 2 Theorv Absorption signal Mode jump Io Cc O D g ES Z D 7 x O O al Figure 2 11 Schematic theoretical comparison between an absorption signal and a mode jump in the direct signal together with their corresponding lock in si
92. out changing anything on the powcr supply for the PMT voltage What does the 10 uA current corresponds to now Remember this value and never let the PMT current exceed this value Lower the PMT voltage to zero Producing a calibration curve for standard addition Follow the next instructions to generate a standard addition curve L 10 Put a piece of polystyrene foam on the sample board in between the fiber and the filter to the PMT Lower the fiber so there is no gap of air between the sample and the fiber but without deforming the sample Raise the voltage slowly over the PMT so the current out from the device is 10 yA Use the value obtained Turn on channel 2 with the modulation signal Move the signal with the injection current to the laser so the signal is centered Press Harm on the lock in amplifier and change it to one This se lects the harmonic studied The first harmonic corresponds to the first derivative of the direct signal and the second harmonic to the second derivative Compare with the preparations Change back to the second harmonic Change the sensitivity setting on the lock in amplifier to achieve an optimal lock in signal e g the highest sensitivity setting achievable without overloading the lock in amplifier note that clicking the down arrow will increase the sensitivity setting Measure and normalize to obtain the GMS signal Raise the fiber 1 cm each time and measu
93. oxygen content in the fruit The ex change of gases in a fruit was also investigated This was done by measuring the change of the oxygen absorption signal over time for a nitrogen exposed tree tomato when placed in air Volcanic rocks Measurements were made on two different types of volcanic rocks pumice and andesite using six samples differing in color and thickness Neither of the rocks gave an oxygen absorption signal This was probably due to the fact that not enough light could pass through the rocks Thinner samples might give more promising results but the reason could also be a lack of molecular oxygen in the samples 8 2 Future work The laboratory at EPN was successfully assembled and was made fully func tional However necessary measurements on reproducibility and accuracy were never made These types of calibrations need to be done to make the laboratory usable for advanced scientific studies T Chapter 8 Summary and conclusions A validation of the system by comparing the GASMAS system with an al ready validated equipment should be done There are facilities at the Depart ment of Food Science at EPN holding equipment for intrusive determination of oxygen content inside matter Contacts have been established with the department to make this possible The validation process is an important part in the development of the GASMAS technique and is a challenging task for future workers on GASMAS at EPN During the entir
94. parts were manufactured in the mechanical workshop at LTH Chapter 1 Introduction MSc Gabriel Somesfalean who had pursued the Harare Project including a visit for integration at University of Zimbabwe also did these arrangements for the project in Quito Although due to his upcoming PhD dissertation the project could not be completed In discussions with Prof Sune Svanberg at the Division of Atomic Physics at LTH a Masters project for the authors was formed with the aim of final izing the original Quito project With the different study background of the authors Electrical Engineering and Engineering Physics useful knowledge could be brought from a wide range to the project MSc Linda Persson graduate student in the Lund diode laser spectroscopy group was involved in the project She functioned as a mentor providing very valuable guidance and inspiration to the project She also made a visit to the Quito site to contribute in the project With the completion of the present project and with the inclusion of an already existing diode laser set up for laser induced fluorescence spectroscopy in Quito being the result of a previous development project 2 diode laser spectroscopy is now firmly established in Ecuador 1 2 Purpose The purpose of the project was to establish a diode laser research and teach ing laboratory at EPN for diode laser absorption spectroscopy In the future this laboratory will allow the possibility to pe
95. pectroscopy Basic Aspects and Practical Applications 2nd ed Springer Verlag Berlin Heidelberg 1992 3 5 G Somesfalean PAD Thesis Environmental Monitoring using Diode Laser Based Spectroscopic Techniques Lund Reports on Atomic Physics LRAP 329 2004 6 J Alnis B Anderson M Sj holm G Somesfalean and S Svanberg Laser spectroscopy of free molecular oxygen dispersed in wood materials Applied Physics B77 691 695 2003 7 O Svelto Principles of Lasers 4th ed Kluwer Academic Plenum Press New York 1998 8 U Gustafsson PhD Thesis Diode Laser Spectroscopy in Extended Wavelength Ranges Lund Reports on Atomic Physics LRAP 253 2000 9 Roithner LaserTechnik Price List sep 2005 Available at http www roithner laser com All_Datasheets Pricelists 10 DFB Laser diodes nanoplus GmbH sep 2005 Avail able at http www nanoplus com content view 23 61 DFB Laser diodes html 84 11 112 13 14 15 16 17 18 19 20 21 22 23 BIBLIOGRAPHY An overview of laser diode characteristics sep 2005 Available at http www ilxlightwave com appnotes overview laser diode _ Characteristics pdf G Somesfalean M Sj holm J Alnis C af Klinteberg S Andersson Engels and S Svanberg Concentration measurement of gas embedded in scattering media by employing absorption and time resolved laser spectroscopy Applied Optics 41 3538 3544 2002 U Gu
96. pendent on the specific backgrounds The project was started January 24th and ended September 7th 2005 During this period there was a break in the hands on work with the equipment between March 6th and May 12th since the equipment was held at the customs of Ecuador Appendix B GASMAS poster GASMAS GAs in Scattering Media Absorption Spectroscopy Marta Cassel Kngquist Christoffer Bjorkwall Linda Persson and Sune Svanberg Atomic Physics Division Lund Institute of Technology P O Bax 118 SE 221 00 Lund Sweden Email Sune Svanberg fysik Ith se ESCUELA POLITECNICA NACIONAL LUND INSTITUTE OF TECHNOLOGY Lund University Edy Ayala and Cesar Costa Departamento de Fisica Escuela Politecnica Nacional Ladron de Guevara E 11 253 Quito Ecuador Email eayala server epn edu ec GASMAS GAs in Scattering Media Absorption Spectroscopy is a novel developed technique for measurements of gas contents inside porous scattering media such as fruit and the body 1 6 This leads to a wide range of applications The key component in the setup is a diode laser which makes it small cheap and easy to handle These aspects are of importance for commercial use in the future So far measurements have only been performed on molecular oxygen However by changing the diode laser and thus the wavelength it is possible to detect other gases such as CO and H O Molecular oxygen concentration measurements inside fruit Molecular oxyge
97. possible differences How is the precision and accu racy of the measurement 13 Rearrange the equipment in the same status as it was when you arrived l3 Appendix E GASMAS laboratory exercise Laboratory exercise GASMAS GAs in Scattering Media Absorption Spectroscopy Christoffer Bjorkwall Marta Cassel Engquist Yolanda Angulo September 2005 Introduction This laboratory exercise will introduce a spectroscopic technique called GAS MAS Scattering absorption and modulation technique will be introduced and studied During the laboratory exercise investigations on polystyrene foam and balsa wood will be done A suitable calibration method standard addition will also be introduced Preparatory problems Please read through the entire laboratory instruction then solve the folowing preparatory problems 1 What could happen if the diode laser gets a too high injection current input If the current output from the PMT gets too high 2 The current from the PMT is measured over an 5 kQ external resis tance What should be the maximum voltage measured on the oscillo scope 3 In Fig 1 there is an example of an absorption signal and a ramp PMT current Calculate the GMS signal The figure indicates the zero level 1 Tek Run kS s Average GK Averages 16 Td Acquisition tode ak Sample ke Peak Detect ix SMS 5 mt Fnvelope 3 Chi SOmV 2V M 50ms Ext Y 260mY Mode Aver
98. re up to 5 cm All together 6 measurements Make a table of GMS vs height in cm and from it make a standard addition curve on the computer and calculate the equivalent mean path length for the sample 16 Measurements on polystyrene foam Different thicknesses Measure and calculate Leg for different thicknesses of polystyrene foam What does the dependency look like Different widths You will be given a set of rectangular pieces of polystyrene foam with differ ent side sizes Measure and calculate Leg for different widths of polystyrene foam How do you think the graph would look like Why do you think you observe a difference in GMS signal for different sizes Explain Measurements on balsa wood Measure and calculate the Leg for a piece of balsa wood Consider the actual size of the sample and compare with the measured Leg Does the actual size correspond to the measured Leg Why or why not Er Appendix F Manual A user s manual named Manual for the GASMAS set up at EPN Quito was made with the following table of contents 10 11 Introduction Background GASMAS Equipment Noise 2 3 4 5 6 7 8 9 The LabVIEW program GASMAS Procedures Frequently asked questions and common problems Appendix A Updating this manual Appendix B Further information
99. rform research and have a good environment for education in the area of absorption spectroscopy Our task was to prepare the equipment in Lund before transport assem ble the experimental stations on site in Ecuador and initiate experimental studies It became clear that there was also a need to perform informational activities about the equipment and its applications for the students and possible cooperation partners on site The purpose of this report is not only to leave a detailed report about the Master s project but also to produce a good source of information for the people who will continue the work on the absorption spectroscopy set up at EPN 1 3 Goal 1 3 Goal The goal of the project was to leave a well functioning absorption spec troscopy laboratory both for education and research together with the know how about it for future researchers and supervisors Another goal was to in vestigate a research field of interest to the country of Ecuador possible to be investigated with the GASMAS technique and to pursue initial experimental studies with the equipment 1 4 Achievements After checking and preparing in Lund the equipment was successfully de livered to EPN after about a two months delay at the customs of Ecuador This time of delay was partially spent by performing information activities such as making a poster holding a presentation and making connections with possible research partners Once the equipment arr
100. s Diode lasers or semiconductor lasers are produced using advanced material processing techniques as a compound of different materials The materials used depend on which wavelength the laser is intended for It is today possible to reach wavelengths between 0 4 to 29 ym The majority of the diode lasers are made of doped materials from group III e g Al Ga In and group V e g N P As Sb in the periodic system Diode lasers made from these materials emit light in the wavelength range 600 1600 nm 7 There are two general types of diode lasers homojunction lasers and heterojunction lasers Homojunction lasers has a more simple construction than heterojunction lasers and will be described as a mean to understand the function of diode lasers Homojunction lasers A homojunction diode laser is created by joining semiconducting materials doped in different ways One part is n doped has an excess of electrons 15 16 Chapter 2 Theory and the other one is p doped which means it has a lack of electrons so called holes When a voltage is applied over the semiconducting material the electrons from the conduction band and holes from the valence band will diffuse and be able to recombine see Fig 2 4 Photons with the energy corresponding to the band gap will be emitted 7 Depletion sy Conduction i band E p region n region puregion n region Valence band Figure 2 4 A homojunction laser with val
101. spersed in scattenng media Optics Letters 26 16 2001 2 G Somesfalean M Sjohoim J Alnis C af Kimteberg S Andersson Engels and S Svanberg Concentration ee TRI COR PE RENE LAND et RNY NTO lacio AQUEL CPOE TF 2002 3 J Alms B Anderson M Sjoholm G Somestalean and S Svanberg Laser spectroscopy of free molecular axygen in wood materials Apphed Physics B 77 691 2003 A sors era EE 5 L Perii E aia M Andersson M Sj holm Studies of gas exchange in fruits using laser spectroscopic techniques FruiictS Submatted 2005 6 L Persson K Svanberg and S Svanberg On the potential of human sinus cavity diagnostics using diode laser gas spectroscopy Appbed Physics B Submaited 2005 Appendix C Presentation abstract GASMAS Applied diode laser spectroscopy Linda Persson Marta Cassel Engquist Christoffer Bj rkwall and Sune Svanberg Atomic Physics Division Lund Institute of Technology P O Box 118 SE 221 00 Lund Sweden e mail linda persson fysik lth se http www atom fysik Ith se Edy Ayala andCesar Costa Departamento de Fisica Escuela Politecnica Nacional Ladron de Guevara E 11 253 Quito Ecuador e mail eayala server epn edu ec GAs in Scattering Media Absorption Spectroscopy GASMAS is a novel technique developed at Lund Institute of Technology by the research group of Professor Sune Svanberg This technique deals with non destructive in situ measurements of gas inside porous materials suc
102. stable and Rb 27 which in principle is radioactive but whose half life is longer than the age of the earth The resonance transitions to the ground state from the lowest excited configuration has the wavelengths 794 7 nm for the D line 5s Sy 2 5p Pi 2 780 2 nm for the D line 5s 281 5p Psja In Fig 1 it is shown how the hyperfine structure looks for the ground state denoted 5s S 2 and the excited state denoted 5p Pyy2 for Rb The en ergy by the different hyperfine structural levels is denoted by the resulting quantum number F resulting from the coupling between the total angular momentum quantum number of the electronic J 1 2 or 3 2 and the nuclear spin I Rb has J 3 2 In Fig 1 the separations in frequency AE h between nearby hyperfine structural levels have been marked out If laser light is to be absorbed the quantum state F cannot be changed arbitrarily For allowed transitions the following rules pertain AP ea 0 ESO to 00 is forbidden Thus we get six spectral lines lying very close to each other in frequency In Fig 2 a recording is shown over the six transitions in Rb The numbers at the transitions in Fig 1 and the peaks in 2 belong together Consider what will happen when the frequency of the laser is increased In Fig 2 even the 3 Figure 1 The energy levels for the rubidium isotope Rb The transitions called D and Da are indicated Transitions between the hyperfine structura
103. stafsson G Somesfalean J Alnis and S Svanberg Frequency modulation spectroscopy with blue diode lasers Applied Optics 39 3774 3780 2000 P Horowitz W Hill The Art of Electronics 2nd ed Cambridge Uni versity Press Cambridge 1989 M Sj holm Diploma Thesis Development of a laser spectroscopic tech nique for gas in scattering media Lund Reports on Atomic Physics LRAP 271 2001 P Kluczynski J Gustafsson A M Lindberg O Axner Wavelength modulation absorption spectroscopy an extensive scrutiny of the gen eration of signals Spectrochimica Acta B56 1277 1354 2001 L Persson H Gao M Sj holm S Svanberg Diode laser absorption spectroscopy for studies of gas exchange in fruits Optical and Laser Engineering in press 2005 L Persson B Anderson M Andersson M Sj holm and S Svanberg Studies of gas exchange in fruits using laser spectroscopic techniques Frutic05 Montpellier 2005 L Persson K Svanberg and S Svanberg On the potential of hu man sinus cavity diagnostics using diode laser gas spectroscopy Applied Physics B in press 2005 The HITRAN database sep 2005 Available at http cfa www harvard edu HITRAN Modified from J L Boulnis Photophysical processes in recent medical laser developtments a review Lasers in Medical Science 1 47 66 1986 Laser Applications sep 2005 Available at http hvperphysics phy astr gsu edu hbase optmod lasapp2 html A V Otieno
104. tandard procedures and measurement methods are described The manual should also be helpful for the trouble shooting of some possible problems with the equipment The manual is written with the intention of constant improvement Hence a section of the manual was dedicated to instruct these updates 5 5 Website A website was set up in order to have a constant and independent source of information about the project This website was put on the EPN web server http www epn edu ec Departamentos fisicaProy html It presents the project gives important contact information and contains the two lab oratory exercise instructions T he website was also constructed in order to be used in the future to present information about ongoing projects etc 43 Chapter 6 Assembly During the preparations in Sweden the entire rubidium set up and the major part of the GASMAS set up were assembled and tested The reason for doing this was not only to test the equipment and to find possible missing parts but to get hands on experience on the GASMAS technique and the set up After arriving into Ecuador there was a two months delay before the equip ment could be cleared from the local customs due to some unexpected prob lems In Ecuador the rubidium set up was mounted first Because of its simplicity this set up could be used to test the other parts of the equipment After getting the rubidium set up to function most of the equipment had ac tually
105. th output and the intensity from a diode laser is dependent on both the temperature and the injection current At low currents the output intensity will be low and the diode will only work as a light emitting diode a LED When the current exceeds a specific threshold current In the diode laser starts to lase see figure 8 This threshold current depends on the temperature By changing the temperature of the diode the band gap and the optical length change The wavelength for GaAlAs type diode lasers changes approximately 0 3 nm C This is the physical reason why diode lasers are possible to tune in wavelength Hence the tunability of the diode laser can be done by changing either or both the injection current and the temperature In this laboratory exercise we will try to keep the temperature constant and use the current to sweep the wavelength Disadvantages There are some disadvantages with diode lasers their output beams are astigmatic assymetric and divergent The astigmatism is a result of the fact that the refractive index has a directional dependence The assymetric and divergent properties of the beam is due to the assymetrical shape of the diode laser normally rectangular 1 ym x 3 um in the active layer This results in a 10 20 x 30 50 divergence see figure 9 The beam however resembles a Gaussian profile which helps to handle this problem Ith Ith2 Ith3 Current Figure 8 The output power as a function of the
106. ties were worked out 1 Prepared a scientific poster and posted it together with other written posters about GASMAS on the walls of the department 2 Performed a presentation about absorption spectroscopy and GASMAS 42 Chapter 5 Technology and knowledge transfer to a general audience 3 Prepared and carried out a laboratory exercise on GASMAS and trans lated an already existing laboratory exercise instruction for rubidium absorption 4 Wrote a user manual for the set up 5 Created a website These five items are described shortly below We believe that as long as people are made aware of the possibilities the chances of making produc tive contacts with other local researchers and of attracting students to get involved grow significantly 5 1 Poster The goal of making a poster see Appendix B was to prepare a single pieced fully comprehensive document of high visual impact describing GASMAS the equipment involved and some possible applications Several copies of this poster were made and posted on the campus along with the invitation to an oral introductory presentation to stimulate the interest of potential attendees 5 2 Presentation In line with our efforts to spread knowledge and inspiration to students and potential collaborate partners at the university an oral presentation was held This presentation was carried out together with MSc Linda Persson from the Lund diode laser spectroscopy group who was visitin
107. tion index of the medium where the light beam travels from and na is the refraction index of the medium being hit by the light beam The reflectance is the fraction of the incident light being reflected 3 2 1 2 Absorption Atoms and molecules absorb energy at certain frequencies or wavelengths This is an effect of their electronic shell structure and the vibrational and rotational energy levels If the photon energy i e its frequency or wave length is suitable to the atom or molecule it may absorb the energy and get excited Every atom and molecule has a unique set of absorption lines i e a fingerprint This makes it possible to identify for example a gas with absorption analysis 4 5 When the absorption is much greater than the scattering it is theoretically described by the Beer Lambert law I v 2 Lupe mer 2 2 It states that the intensity of the incident light J is attenuated exponentially through an absorbing material see Fig 2 2 The cross section of absorption g is the probability of absorption with a unit of area per molecule or atom The concentration of absorbing molecules or atoms is c and z is the length traveled through the medium When no scattering occurs this length is the same as the physical thickness of the sample The cross section is frequency dependent matching the energy level structure 4 5 A derived property a v is called the absorbance It is defined as a om a 2 3 where o v
108. together with the interval 6 3 LabVIEW 09 between each measurement and starts the measurement The user interface during run time can be seen in Fig 6 6 Last saved oscilloscope graph A ce j 32 57 00 062 PM 3 57 00 331 PM 3 8 17 2005 8 17 2005 ary f pes peta wart MESEDE 3 he dd Menage Number o 241 co Tine to next messurement sec Cancel Measurement 3136 Canc Figure 6 6 The user interface of the LabVIEW program during run time for the function Continuous measurement Chapter 7 Experimental work Experimental work was only performed on the GASMAS set up Different experiments were performed to test the functionality and show the possibili ties of the system The first primary test to verify the absorption signals was a lengthy procedure Before measuring on real samples different standard addition curves on atmospheric gas were measured to be able to calculate the equivalent mean path length for different samples With the standard addition curves it was also possible to measure how much molecular oxygen existed intrinsically in the set up This was due to small spaces of air in between the optical surfaces in the light path This resulted in a signal offset that one had to take into account while performing measurements Measurements on polystyrene foam slabs were made to study the dependency of the equivalent mean path length on the physical size width and height of the scattering medium Ad
109. troduction Before the beginning of this project there were only two laboratories in the world with GASMAS set ups GAs in Scattering Media Absorption Spectroscopy 1 The original set up is located at the Division of Atomic Physics at Lund Institute of Technology LTH in Lund Sweden Another one donated by International Science Programme and assembled by the Lund diode laser spectroscopy group at LTH is located at the University of Zimbabwe in Harare Zimbabwe As a result of this Master s project there is now also one at the Department of Physics at Escuela Polit cnica Na cional EPN in Quito Ecuador This report will describe the full process of preparing mounting and testing a diode laser spectroscopy laboratory containing both a GASMAS set up and an absorption set up for rubidium gas from its beginning in Lund to its conclusion in Quito 1 1 Background In 1961 an organization named International Science Programme ISP was founded at the Uppsala University Sweden in order to improve research in developing countries in mathematical physical and chemical sciences Its philosophy is to help build up research on site in the countries in long term cooperations After an application from Prof Edy Ayala at EPN the orga nization decided in 2004 to finance a laboratory for absorption spectroscopy research at EPN The Division of Atomic Physics at LTH was made respon sible for ordering the equipment The equipment was bought and
110. ulation frequency fref and the other one is our detector signal faet Both of the signals are mixcd in the lock in amplifier and the result is fres Jde Since both of 9 the signals have the same frequency the result will be a DC level fref faet which will be amplified and a high frequency signal fref fact which will be filtered away The DC level depends on the phase relation between the two incoming signals and their amplitudes It is possible to adjust the phase relation on the lock in amplifier so that a maximum DC level can be achieved The detector signal and reference signals are then in phase with each other To sum up it can be said that the lock in amplifier works like a filter stops all signals not marked with the modulation frequency fref and amplifies signals modulated at this frequency Procedure Fig 6 shows schematically how the equipment will be connected and what your experimental set up will look like after you performed the tasks 1 to 10 i Temperate gerermar conto fr Cir Nec anakin amp hor ei r 2 egal carte Tes uh Cee ar Cage laser dise disdr Figure 6 A schematic of the experimental set up used in the absorption measurement l First get accustomed to the function generator by connecting it to the oscilloscope and make sure you can generate a saw tooth wave A suitable ramp frequency is 100 Hz and a suitable voltage amplitude is about 40 mV peak to peak value W
111. ummary and conclusions Ghana We would also like to thank ISP for not only sponsoring the whole project itself but also our plane tickets to Ecuador Thank you Lennart Hasselgren and AnnaKarin Norling also for all administrative help We would like to thank Applets Hus at Kiviks Musteri for a scholarship We hope you will have use for the GASMAS technology in the future Our thanks also goes to Mats Lundqvist for helping us getting started with Latex the program this report is written in and Olle Carlsson at ThorLabs bringing the words Product Support to a new dimension We also thank Ake Bergquist for preparing the equipment for the Ecuatorian line voltage system and our Spanish teachers with whose help we could actually manage to speak Spanish in the end Last but not least we would like to thank family friends and our beloved ones Vannesa and Magnus Thank you for understanding the need of work ing far away and long hours A special thanks to Ake the father of Marta who with his visit to Quito shared different ideas and guidance Bibliography 1 M Sj holm G Somesfalean J Alnis S Andersson Engels and S Svan berg Analysis of gas dispersed in scattering media Optics Letters 26 16 18 2001 2 S Svanberg Laser spectroscopy in development EPN Europhysics News 2001 3 F L Pedrotti S J Pedrotti Introduction to Optics 2nd ed Prentice hall New Jersey 1993 4 S Svanberg Atomic and Molecular S
112. vels through an absorbing material see figure 11 In v 2 ue ez 1 The absorption cross section a is the probability of absorption in the unit area per molecule or atom The concentration of absorbing molecules or atoms per volume unit is c and z is the length traveled through the medium The cross section and thus the transmitted intensity are both highly fre quency dependent matching the energy level structure 8 Intensity x Length Figure 11 The absorption of a beam with intensity lo Absorption spectroscopy An absorption spectroscopy set up consists of three main parts a light source a sample and a detector The light is sent through the sample and the output light is detected and measured as shown in figure 12 There are different types of absorption spectroscopies We will focus on the one used by GASMAS ENSA AE N TTE A CI eee x ae a RAS DIS e Light source Absorbing material Detector Figure 12 The three main parts of an absorption spectroscopy set up for gas samples The properties which can be achieved through absorption spectroscopy is the concentration temperature and pressure of the sample The concentration can be calculated from the Beer Lambert law equation 1 by measuring Ir and Jo see figure 13 and knowing the path length and the absorption cross section The light source In GASMAS a diode laser is used as a light source of the laser light By letting the injection current have
113. vestigate papaya potato and guayaba but the signals are not ideal with respect to signal to noise ratio making possible measurements slightly unreliable The reason for why the papaya showed a signal but not the smaller Hawaiian papaya could not be determined It is difficult to evaluate the equivalent mean path length measured on the majority of these fruits and vegetables since they have never been tested before with this technique However there has been measurements on apples before 17 18 and the value obtained in our experiment is substantially larger The reason for this could not be determined but may be related to a difference in the fruits 7 5 Cutting a papaya There are two types of papaya available on the markets in Quito The smaller one is called Hawaiian papaya and the other larger one is solely called papaya The larger papaya is cheaper per kilogram than the Hawaiian It is common to make juice from the larger papaya but the juice is slightly sour A common trick to sweeten the papaya is to cut it either in superficial stripes on the skin or cut the ends off and leave it for the night making the sour liquid leak out By doing this the papaya gets much sweeter 7 5 1 Method First the oxygen absorption signal was measured in a fresh papaya for about 30 minutes Thereafter long stripes were cut in the papaya see Fig 7 6 so that there would be a contact between the pulp and the air The measure ment of the oxyg
114. what happens with the absorption from rubidium if you place the grey filter with 100 times damping in the beam path before and aftcr the absorption cell Try to explain the difference Make sure gi that the absorption is linearly dependent on the laser intensity 8 Investigate and cxplain what happens with the absorption signal if you change the injection current to the diode laser and the voltage amplitude on the function generator respectively Adjust the injection current and the voltage amplitude so that the largest possible part of the wavelength sweep for the absorption signal is being used 9 Split off laser light from the beam path before the rubidium cell and aim it through the Fabry P rot etalon The laser light should hit a detector with a built in amplifier after the etalon connected to the oscilloscope When the reflecting surfaces of the etalon are perpendicu lar to the laser beam interference fringes can be observed The angle of the etalon in relation to the laser beam can be fine tuned with the help of two alignment screws on the etalon holder The length of the etalon is imprinted on its side The index of diffraction of the etalon glass is 1 51118 BK7 and 1 51075 K8 respectively at the wavelengths of interest Thus it is easy to calculate the separation in frequency between the interference fringes the free spectral range 10 Now save the oscilloscope signal from the detector both from Rb ab sorption a

Open Access

Contents

Download Pdf Manuals

Related Search

Related Contents