CO2 laser heating system for thermal compensation of test masses in high power optical cavities. Submitted by: SHUBHAM KUMAR to Prof. DAVID BLAIR
Abstract This report gives a description of the setting up of a heating arrangement using a CO2 laser, which is a part of the planned thermal compensation system to be set up at the south end-station of the AIGO research facility at Gingin. Detailed in this report is the description of the CO2 laser which is used for the heating of the test mass and the electronic feedback stabilization system made for stabilizing the output power of the laser. 1. Introduction The laser interferometer being developed for the detection of gravitational waves uses a resonance cavity in between two test-masses for recycling the laser beam power. This is essential for greater power build up in the cavity and therefore better sensitivity of the interferometer. The power resonating capacity in the cavity is of the order of a few kilowatts, which is high enough for the heating of the test mass causing a significant thermal lens. This thermal lens is formed by the following mechanism: the medium is hotter on the beam axis, compared with the outer regions, typically causing some transverse gradient of the refractive index. The thermal lensing causes distortions in optical components and changes the mode quality of the optical beam, severely impairing the sensitivity of the interferometer. So, there is the need for a Thermal Compensation System to compensate for the thermal lensing distortions. We aim to compensate thermal lensing by using a laser to heat the test mass such as to create a negative lens, which neutralises the effect of the lens created by the laser in the recycling cavity. The heating arrangement consists primarily of a CO2 laser shining on the ETM, and heating it in order to compensate for the thermal distortion caused by the absorption of the laser power recycling in the cavity.
2. Experimental design The components in the arrangement consist of: 1. A CO2 laser for the heating of the test mass. 2. Feed-back stabilisation circuit, for stabilising the CO2 laser output power. 3. He-Ne laser for guiding the invisible CO2 beam through the viewport onto the test mass. 4. An arrangement of suitable lenses and mirrors for increasing the spot size of the CO2 laser beam to a suitable size at the test mass. The diagram of the arrangement is shown below in Fig. 2.1:
Fig. 2.1: The design of the optical table for the setup of the heating arrangement of the TCS. 3. The CO2 laser 3.1 Characteristics: For the purpose of heating the test mass, the most efficient laser is CO2. This is because CO2 is readily absorbed by the optical material being used as the test mass. The CO2 laser is a gas type laser with a wavelength of 10.6µm in the invisible, infrared region. The CO2 laser we are using is a low cost laser of Chinese manufacture. The operating voltage applied to the laser is 3-5 Volts, and the power output of the laser is in the order of a few Watts. 3.2 Study of laser beam quality 3.2.1 Description The beam profile was measured using a beamscan(a rotating-slit type) and the following observations were made: a. The beam quality was poor at low voltages and got increasingly better with increase in voltage. b. The beam showed a fringe pattern, in a manner such as superposed with a Gaussian curve. This could be due to some interference arising from misalignment of mirrors inside the laser tube. The beam pattern at a distance of 250mm from the laser is shown below in Fig.3.2.1
Fig.3.2.1: The beam profile at a distance of 250mm from the laser, X and Y axes respectively. The X-axis is the beam width in micrometers, and the Y-axis is the relative beam intensity. c. The beam quality is dependent upon the distance from the laser. It is very poor, close to the mouth, and gets slightly better we move further away from the mouth. The profiles at different distances are shown in Fig.3.2.2 below. (a)
(b) (c) Fig. 3.2.2: Comparing the beam quality at different distances from the laser (a)50mm (b)350mm (c)450mm. The quality becomes slightly better further away from the laser. The X-axis is the beam width, and the Y-axis is the relative beam intensity. d. The beam profile data is calculated by a beamscan and the values are given below in Table 1: Distance from laser(mm) Spot size of beam Average X- value(µm) Spot size of beam Average Y- value(µm) 0 6525 7140 50 3976 3457 100 3812 3551 150 3817 4454 200 3916 3918
250 3584 3511 300 3958 4075 350 4246 4075 400 4314 4401 450 4624 4353 500 4620 4395 Table1: Beam profile data as calculated by a rotating-slit beamscan. 3.2.2 Results: The data is analysed using a matlab program gauss633 which determines the waist-position of the beam as well as its waist size.
The program calculates the profile of a Gaussian beam of a particular wavelength, by taking in as input the spot size of the beam at various intervals. The graphs generated by the program are shown in the Fig.3.3.1 below: Fig.3.3.1: The graph of the beam profile as calculated by the program gauss633 for X and Y axes respectively. The circles represent the actual experimental data, while the smooth curve is the beam profile as calculated by the program gauss633.
The value of the waist position and waist size is given in table 2 below: X-axis Y-axis Waist position -310.64 mm -293.21 mm Spot-size at waist 1.525 mm 1.91 mm Table2: The values for the waist size and position, as calculated by gauss633 program. 3.2.3 Conclusions 1. The laser is a low quality one and has a poor beam profile. 2. There is the presence of fringes in the profile which could be a result of poor mirror alignment. 3. The waist position and size is calculated and shown in Table2. 4. The laser can still be used for heating purpose, since the heating will not be affected by the high frequency fringe pattern in the profile.
3.3 CO2 Laser profile upon introduction of a lens 3.3.1 Description We now want to study the laser beam profile upon introduction of a lens. The arrangement is shown in Fig. 3.4.1. CO2 laser d x Lens(f) Beamscan Fig.3.4.1: Diagram of the setup for measuring the beam profile upon introduction of a lens. It is essential for us to know the behaviour of the beam when passed through the various lenses used in the whole arrangement, in order to estimate the beam spot size on the test mass during the heating. We measure the beam size at different locations from the laser using the same beamscan as in the previous measurement of CO2 profile. Similarly we make use of the matlab program gauss633 to calculate the waist location and waist size of the beam. The data for this is shown in Table 3, and the graph for the beam size variation with distance, as generated by the program gauss633 is shown in Fig. 3.4.2. Distance of beamscan (x) Spot-size of beam. Average X-value (µm) Spot-size of beam. Average Y-value (µm) 200mm 3672 3741 250mm 3513 3250 350mm 2783 2738 450mm 2687 2750 500mm 2632 2587 550mm 2887 2896 650mm 3648 3507
750mm 4741 4462 900mm 6110 5844 Table 3: Beam profile data for the setup with a lens introduced. We also made use of another matlab program, Alexei s code, to predict the variation of the spot size of a Gaussian beam when passed through a lens or a combination of lenses. Through this program we created a model of the optical table used for the whole setup. The graph of the beam propagation after passing through a single lens, as predicted by the program, is as shown in Fig.3.4.3 We can compare the two profiles; one as derived from the program gauss633 using data measured by the beamscan, and the second as predicted by the program Alexei s code. The waist position and size from the two graphs are compared in Table 4. Fig. 3.4.2: The graph of the beam profile calculated by gauss633 taking the data from Table3. For X and Y axes respectively. The circles represent the actual experimental data, while the smooth curve is the beam profile as calculated by the program gauss633.
Fig. 3.4.3: The graph of the beam profile as predicted by the program Alexei s code. The X-axis gives the distance from the laser(in mm) and the Y-axis is the beam radius. Measured value (X Alexei s code only) Position of waist 500mm 500mm Size of waist 1316µm 750µm Table4: Comparison between the values of the graphs of Fig. 3.4.2 and Fig.3.4.3. This comparison gives an idea about the accuracy with which we can use the program for further calculation of the beam spot-size, upon introduction of more lenses and mirrors. 3.3.2 Conclusions From the comparison of the two profiles the following conclusions can be drawn: 1. Both the profiles show the same position of beam waist, which is at 500mm from the laser. 2. There is discrepancy in the waist size of the two profiles which needs to be looked into. Due to time constraints I am not going to looking further into this discrepancy.
4. Stabilization of CO2 laser The CO2 laser used is very unstable with significant fluctuation in intensity with respect to time. This fluctuation makes the laser unacceptable for heating purpose as the heating will become uncontrolled due to the laser instability. Therefore, a system for the stabilization of the CO2 laser is absolutely necessary. We have to keep in mind that it is more important to control low frequency fluctuations than the high frequency ones, since the high frequency does not disturb the heating of the test mass very significantly unlike the low frequency ones. We designed an electronic feedback circuit to do the laser stabilization. The plan of the circuit was made by Dr. Chunnong Zhao. 4.1 Description The basic plan of the stabilization system is: 1. A beam splitter picks up a very small part of the beam and focuses it onto an infrared sensor. 2. The signal from the sensor is passed through an OpAmp. which acts as an offset control for the signal, giving it an offset which can be tuned desirably.
3. The signal from the offset is input to a low pass filter, and the output is then amplified and an offset can be added, up to 5 volts. 4. This output is then fed back to the laser. The schematic for the circuit is shown in Fig. 4.1 S1 S2 Fig. 4.1: Schematic for the feedback stabilization electronic circuit. 4.2 Details of the circuit: 1. Components used: P1 - Infrared sensor R1, R2, R4, R6 - Resistance (1 kω each) R3 - Variable resistance () R5- Resistance (10 kω) R7- Variable resistance R8- Variable resistance C1 - Capacitor (0.16 µf) VR1, VR2 - Voltage regulator (LM7805)
OP1, OP2, OP3 Operational amplifier (OP07) S1,S2 Switch 4.3 Function: The circuit can be broken up into 3 parts, arranged in series: a. The initial offset control: The voltage regulator VR1 supplies a constant +5Volts. By varying the variable resistor R3, we can change the voltage input into the positive terminal of the OpAmp (OP1). The OpAmp then subtracts the input signal (from the infrared sensor P1), from this voltage. By adjusting the variable resistor R3, we can change the output voltage from the OpAmp. to become very close to zero. This is important for the next part of the circuit, into which the output from OP1 is fed. b. Low pass filter The next part of the circuit is a low pass filter with a pass frequency f0 =100 Hz. The formula for calculating this f0 is: fo = 0.16/ (R5C1) So, for fo =100Hz, and R5= 10 kω, we have C1= 0.16µF The output signal is then fed into the third part of the circuit for amplification and offset adjustment. 3. The final offset control This part is similar to the first part of the circuit. The VR2 supplies +5 Volts, which is changed by the variable resistance and this voltage (say +V1 volts) is fed into the positive terminal of the OpAmp, while the signal from the filter is fed to the negative terminal. The OpAmp then subtracts the signal from the +V1.
By tuning the variable resistor R7, we can change the gain of the OpAmp and thus the voltage output. By carefully tuning the variable resistors R7 and R8, we can vary the Gain of the circuit and the voltage input to the laser, which is optimum at 5 volts. This voltage drives the laser. This is therefore the feedback loop which stabilises the laser. 4.4 Testing the circuit The circuit was tested with the CO2 laser and the beam intensity was recorded on an oscilloscope. The graph of intensity versus time for in-stabilized and stabilized laser is shown in Fig. 4.4.1and 4.4.2 respectively. (a) (b) Fig. 4.4.1: The oscilloscope image for the unstable CO2 laser. The time duration of the graph is 1000s for each graph. The top curve shows the power output from the laser, while the bottom line is the power input to the laser (about 2V). The scales are different for the two curves. Both (a)and (b) show the instability of the laser and represent two different set of recordings.
(a) (b) Fig. 4.4.2: The oscilloscope image for the stable CO2 laser. Time duration is 1000s for each graph. The top curve shows the power output from the laser, while the bottom line is the power input to the laser (about 2V). (a)the middle line is to monitor the signal in the circuit after the initial offset; it should and it is, almost zero. (b)the reason for the sudden fluctuation is not known presently, although it seems that it is due a slight mechanical disturbance to the gain knob of the control. 4.5 Conclusions The following conclusions can be made from the graphs of the laser intensity output variation with respect to time, before and after stabilization: 1. The instability of the laser is considerably high, as is evident from the graphs, so there is an absolute necessity for this stabilization system. 2. There is an oscillation of 50Hz that is being coupled into the detector and therefore in the input signal to the circuit. The
coupling is from an external source although where exactly from is not yet known. 2. The circuit works satisfactorily for stabilizing the laser. 3. There was a sudden oscillation as shown in Fig. 4.4.2(b), the cause of which is not known yet. 4.6 Suggestions for further improvement of the circuit The circuit can be further improved by separating the gain stage from the final offset stage and making them as two independent stages. At present when we change the gain of the circuit, we have to readjust the offset too, but by separating the two stages this readjustment won t be needed, thereby making it more convenient to operate. 5. The optical table design 5.1 Design The final design of the optical table is shown in Fig.5.1. It takes into account the following factors: The beam spot-size should be between 10-20mm at the test mass.
A He-Ne laser is used to direst the CO2 beam, so both the beams have to be perfectly aligned with each other. A dummy test-mass is initially used to see the effects of the heating, so this must be at the same distance from the laser as the actual test-mass. Fig. 5.1: The design of the optical table for the setup of the heating arrangement of the TCS. The optical table has been installed along with the laser stabilization circuit. The photograph of the whole arrangement is shown in Fig. 5.2 below:
(a) (b) Fig.5.2: Photos of the actual setup. (a) The optical table; the glass tube is the CO2 laser, while the black tube beside it is the He-Ne laser. (b)the stabilization system, comprising of the stabilizing circuit, a 12V DC voltage supply, a photo-detector and an oscilloscope for monitoring.
5.2. Conclusion The optical table is ready and installed in the laboratory. Work is ready to begin on the dummy test mass, before the actual test mass in the laser resonator cavity. 6. Acknowledgements This is the report of the work done by me during the two months internship training at AIGO, Gingin. I am thankful to my Professors David Blair and Chunnong Zhao for continuously helping me throughout this project and going through the report over again. I also thank Sunil Susmisthan who worked along with me on this project. I also thank PhD students Pablo and Jean-Charles, and Andrew who were always there for assistance whenever i needed any.