Interference lter stabilized external cavity diode laser Matthias Scholl March 18, 2010

Abstract This document contains explanations about an external cavity diode laser (ECDL) using an interference lter as the wavelength selective element as well as the characterization of two nearly identical realizations. Spectral properties and characteristics are measured and discussed. At the end a step by step instructions manual for building one of these ECDL's is given.

1

Contents

I Theory and Measurements

4

1

4

2

Introduction and Motivation 1.1

Littrow conguration . . . . . . . . . . . . . . . . . . . . . . . . .

4

1.2

Interference lter conguration

5

. . . . . . . . . . . . . . . . . . .

Basic Setup and Cavity Modes

6

2.1

Elements used in the cavity

. . . . . . . . . . . . . . . . . . . . .

6

2.2

Internal & external cavity modes . . . . . . . . . . . . . . . . . .

12

3

Output Power

15

4

Beam prole

16

5

Wavelength tuning

18

6

Noise and linewidth

20

6.1

Denition: Linewidth, Drift and instantaneous linewidth . . . . .

21

6.2

Contribution of driver noise to linewidth . . . . . . . . . . . . . .

21

6.3

Noise spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . .

22

6.4

Beat note measurements . . . . . . . . . . . . . . . . . . . . . . .

22

6.5

Piezo resonance frequency . . . . . . . . . . . . . . . . . . . . . .

25

II Instructions for building the ECDL

29

7

Schematic of the laser

30

8

9

Needed items

30

8.1

Drivers & controllers . . . . . . . . . . . . . . . . . . . . . . . . .

30

8.2

Mechanical parts . . . . . . . . . . . . . . . . . . . . . . . . . . .

30

8.3

Optics & electronics

. . . . . . . . . . . . . . . . . . . . . . . . .

31

8.4

Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

31

Cleaning of mechanical parts

32

10 Wiring - Electric Connections 10.1 Thermal management wiring

32 . . . . . . . . . . . . . . . . . . . .

33

10.2 Wiring for laser diode current . . . . . . . . . . . . . . . . . . . .

34

10.3 Wiring for the piezo

35

. . . . . . . . . . . . . . . . . . . . . . . . .

11 Assembling: out-coupler to piezo

35

12 Assembling: piezo+out-coupler to cat's eye mount

36

2

13 Soldering and testing: socket 13.1 Soldering

peltiers, thermistors and laser diode 37

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39

13.2 Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39

14 Assembling: cavity to base plate

41

14.1 Cavity, peltier and base plate . . . . . . . . . . . . . . . . . . . .

41

14.2 Thermistor for base

42

. . . . . . . . . . . . . . . . . . . . . . . . .

15 Assembling: laser diode to copper housing

42

16 Assembling: laser diode+copper housing to cavity

45

16.1 Preparation: Cat's eye mount to cavity 16.2 Assembling

. . . . . . . . . . . . . .

45

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

45

17 Optical setup for testing and optimization

46

18 Collimation lens

47

19 Cat's eye lenses

48

19.1 Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

48

19.2 Observing the threshold

49

. . . . . . . . . . . . . . . . . . . . . . .

19.3 Optimizing the lens positions

. . . . . . . . . . . . . . . . . . . .

49

20 Interference lter

51

21 Saturation spectroscopy

52

21.1 Finding the Potassium line - Fluorescence . . . . . . . . . . . . .

52

21.2 Observing saturation spectroscopy signal . . . . . . . . . . . . . .

52

21.3 Optimizing external and internal cavity modes

53

. . . . . . . . . .

22 What to improve

53

23 Costs for one laser

54

A Photos of Mechanical Parts

56

B Beam prole pictures

60

3

Part I

Theory and Measurements 1

Introduction and Motivation

Laser diodes are an attractive light source in many elds of experimental physics and technology. They combine the advantage of being very small (smaller than a thumb) and due to the industrial use in CD/DVD/Blue Ray, being quite cheap (at least for the wavelength region around 780nm, 767nm and 405nm, which are important for laser cooling applications with Rb or K). Output powers vary from

µW to several Watts.

The disadvantage of laser diodes are their bad spatial

mode prole, usually elliptical with dierent higher TEM modes, and the big linewidth. Often laser diodes operate in multi mode, meaning they emit laser light at more than one frequency at the same time. Since many applications, e.g. laser cooling, need a single laser frequency and a linewidth on the order of a few MHz and lower, people have developed several techniques to force a laser diode to single mode operation and to narrow the linewidth. One is to use an external cavity, to introduce frequency dependend additional losses to the laser, in order to discriminate most of the modes, so that the gain for one particular mode is much higher than for all other modes. Due to a certain frequency width of this discriminating element, the linewidth can also be reduced.

1.1 Littrow conguration The most common used external cavity design is called Littrow conguration (Figure 1). In this setup, the 1st order diracted light from a diraction grating gets fed back to the laser diode and the 0th order light is used for out coupling. Since the angle of the 1st order diracted light depends on the frequency of the light, only a very narrow frequency band gets fed back to the laser and therefore the losses for all other frequencies are much higher. By turning the diraction grating, one can tune the frequency that gets fed back and thus the wavelength emitted by the laser. The wavelength dependence of the 1st order angle is given by the Bragg

λ = 2d · sin θ with θ the angle of incidence and d the grating constanst. d−1 = 1200 lines/mm and θ ≈ 28° for a wavelength of 780nm. leads to a wavelength sensitivity of dλ/dθ ≈ 1.5 nm/mrad for turning the

condition

Typical values are This

grating. While turning the grating, the 0th order out coupling beam also moves, which redirects the beam by

(dλ/dθ)−1 ≈ 1.3 mrad/nm.

In this conguration of an external cavity, the amount of feedback and the wavelength selection, since performed by the same element, are not independent, which results in less control over mode hops and mode hop free scanning range.

4

Figure 1: Littrow conguration, using a diraction grating for an external cavity

1.2 Interference lter conguration In 2006 Clairon et al. presented a new design of an external cavity (Figure 2) [Clairon] using a narrow-band interference lter (IF) for wavelength discrimination, instead of using a diraction grating. The interference lter has 90% transmission and a FWHM of 0.3nm. The wavelength of the transmission maximum depends on the angle of incidence (details see 2.1.2). Therefore the wavelength tuning can be performed by turning the lter. This leads to a sensitivity of

dλ/dθ ≈ −0.017 nm/mrad,

which is 60 times smaller than for the Littrow

conguration. Thus the laser design is in principle less sensitive to mechanical vibrations and disturbances. The feedback is performed by a partial reector, which can be displaced by a piezo tube. Still the beam gets displaced while tuning the wavelength, but the displacement is reduced by a factor of 2 compared to the Littrow conguration. Also the independence of the wavelength selection, using the IF and the optical feedback, using the out coupler mirror (OC) is an advantage of the new conguration. In the following sections a more detailed explanation of the laser design, the spectral properties and mode dynamics is given.

5

Figure 2: New conguration of the external cavity, using an narrow band interference lter for wavelength discrimination. The focal lengths and reectivities are given for our specic case.

2

Basic Setup and Cavity Modes

2.1 Elements used in the cavity In this subsection details are given about properties and purpose of the different elements in the laser setup (shown in Figure 2), both theoretically and experimentally.

2.1.1

Laser diode

The laser diode is anti reection coated on the out put facet, which leads to a larger total tuning range of the wavelength, using an external cavity (see [Rasel]. Due to the AR-coating on the laser diode, there is just a weak resonator in the bare laser diode. But it is interesting to look at the characteristic curve to nd out, if the bare diode still can start lasing without additional feedback from an external cavity and what the threshold current is. To do this, the diode was mounted to a Thorlabs laser diode mount and the emitted light was collimated and send to a power meter.

By looking at the

output power for dierent laser diode currents one obtains the characteristic curve shown in Figure 3. The threshold current is tted to be

Ithr = 65mA.

The total tuning range of the laser (using the IF rotation) is determined by the gain curve of the laser diode, shown in Fig 2.1.1. It basically reects the properties of the laser transition in the band structure of the GaAs semiconductor, the diode is made of.

2.1.2

Interference lter

The IF takes the part of the wavelength discrimination in the external cavity, meaning it is the part in the resonator which forces the laser to stable single

6

Figure 3: Bare laser diode characteristic curve.

Figure 4: Laser diode gain curve [datasheet]

7

center wavelength

780nm

767nm

angle of incidence

≈5° ≈ 0.017 nm/mrad

≈22° ≈ 0.071 nm/mrad

wavelength sensitivity

Table 1: Important properties of the interference lter

mode and reduces the linewidth. It is also used for coarse wavelength tuning. It works on the basis of the interference of multiple reected beams, similar to a fabry perot interferometer. The lter is made of a substrat, which is coated with many dielectric layers on one side and anti reection coated on the other side. The blue shimmering side is the AR-coated side, because all the red light gets transmitted, but the AR coating is very bad for blue light, which gets reected. The other side is shimmering red, since only a narrow band in the red light frequency region gets transmitted, the rest is reected. Since the laser is planned to run at 767nm, it is important to know, how the wavelength of the transmission maximum depends on the angle of incidence. With that information we can nd out at what angle the lter should be used for 767nm and what the sensitivity

dλ/dθ

is. This quantity gives information,

how sensitive the system is to vibrations or misalignments. Theoretically the lter can be described as a thin fabry perot etalon with an eective index of refraction

neff .

The transmitted wavelength is given by:

s 1−

λ = λmax where

θ

sin2 θ n2eff

is the angle of incidence (AOI) and

λmax

(1)

is the transmitted wave-

length at normal incidence. The center wavelength vs aoi of the lter (Figure 5) is measured using a white light source. The light gets collimated and send through the lter, which is mounted to a rotation stage.

The transmitted light is spectrally analyzed

with a monochromator. The resolution is 0.1nm, limited by the width of the monochromator slit. The data can be tted very well using Equation (1).

λmax is measured to neff = 1.97 ± 0.01,

780.9nm and the eective index of refraction is tted to be which is a typical value for

neff .

With this result, one can nd the AOI for 767nm and for 780nm.

The

wavelength sensitivity can be calculated using the derivative of Eq (1) . The results are shown in Table 1 : The sensitivity at 780nm matches the value in [Clairon], but at 767nm the lter, and therefore the laser, is 4 times more sensitive! The FWHM of the transmission maximum at AOI=0° is measured to be

0.3nm ± 0.1nm. Figure 6).

By changing the AOI, the FWHM seems to get bigger (see

But this eect can be explained by assuming a not perfectly col-

limated beam: An uncertainty

∆θ

in the angle of incidence, translates to an

8

Figure 5:

The maximum of the transmitted wavelength spectrum is plotted

versus the angle of incidence. The AOI is changed, using a rotation stage as the underlayment of the lter.

9

Figure 6: FWHM of the transmission maximum in dependence of the angle of incidence

uncertainty of the center wavelength

∆λ = |

λ(θ)

of

d λ(θ)| · ∆θ dθ

(2)

and therefore to an eective broadening of the transmission curve. In a sim-

∆λ added to the intrinsic neff = 1.97, the data can be tted The uncertainty in the AOI is tted to be: ∆θ = 0.7° ± 0.1°. This agrees

ple model, the total measured linewidth is given by FWHM@0°. Assuming nicely.

λmax = 780.9nm

and

with the measured divergence angle of the white light beam. So as a result, the FWHM of the lter itself of 0.3nm does not change signicantly, when turning the lter. The transmission at 767nm has been measured to be

2.1.3

90.7% ± 0.3%.

Lenses

In the laser setup three lenses are used, all with an anti reection coating: 1. A Collimation lens with f=4.51mm and high NA=0.55 is used to collimate the light coming out from the laser diode.

Laser diodes emitt strongly

divergent light compared to other laser sources, with an elliptical beam prole (divergence angles of about 30° parallel and 10° perpendicular). The collimation lens is necessary, to collect as much light from the diode

10

as possible (high NA). But a good collimation is also needed in order to reduce the additional broadening of the FWHM of the interference lter (see 2.1.2). 2. The optical feedback is provided by a combination of a lens (L1) with f=18.4mm and the out-coupler mirror (OC) in focal distance.

This so

called cat's eye conguration is less sensitive to misalignments of the OC, compared to the case of feeding back without having this lens. Especially for the case, that the OC is slightly tilted (which will usually hold in lab reality), it is easy to see that the cat's eye conguration provides a better feedback. 3. The outcoupled beam gets collimated again by a lens with f=11.4mm. This lens has nothing to do with the quality of the external cavity. It is just needed to provide a nice collimated output beam, to use the laser light for experiments, measurements, ber coupling, etc.

The focal length of

L2 is chosen to be smaller than the focal length of L1, in order to decrease the size of the output beam, since it is more convenient to work with a smaller beam (easier ber coupling, less problems with apertures, etc).

2.1.4

Out coupler & piezo tube

The out-coupler mirror, which provides the feedback, together with the lens L1 (cat's eye conguration) is glued to a piezo ring actuator. The out-coupler has a diameter of 0.500, a thickness of 0.125 and is at. It is partial reecting on one side and AR-coated on the other side (both made for 767nm). The reectivity of the OC determines the amount of feedback and thereforce eects the output power of the laser as well as the instantaneous linewidth.

Higher reectances lead to higher intracavity powers, but less out

coupled (usable) light. Considering the external cavity as a fabry perot interferometer/etalon, higher reectances give a higher nesse and thus narrower cavity modes, meaning a narrower instantaneous linewidth. Lower reectances however, lead to a lower nesse, but more out coupled light. But if the reectance is too small, the external cavity gets too weak and the laser properties are given by the bare diode again. This means, that there is a point at which, decreasing OC reectivity leads to a decreasing output power again! At some point, we expect a nice tradeo between linewidth and output power, that satises the needs for our laser cooling applications. To nd this point we characterized identical laser systems with OC reectivitys of R=8%, R=11%, R=17% and R=22%. The ring actuator has a length of 9mm and a diameter of the isolation coating.

&4.5mm

≈12mm, including

The diameter of the cylindrical hole in the middle is

including the isolation coating.

As any other solid state body, a piezo has a resonance frequency. It theoretically follows a simple spring-mass-system, according to the information on www.piezomechanik.com:

11

f0 = where S is the stiness and

1 2π

meff

r

S meff

(3)

is approximately 1/3 of the mass of the

ceramic + the mass of all installed pieces. For our particular piezotube the stiness is

S = 200 N/µm and for a one side = 40 kHz [Datasheet].

x oscillation, the axial resonance frequency is given as f0

The goal is now to calculate the resonance frequency of the piezo xed on one side and with a mirror glued on the other side. Since

f0

and S is given we can calculate

meff ≈ 3g

for the piezo without

mirror.

mOC ≈ 1g, using the 2.5 g/cm3 ). + mOC ≈ 4g and thus the

The mass of the out-coupler mirror can be calculated to

dimensions of the mirror and the density of the substrate BK7 (≈ Therefore the new eective mass is

0

meff = meff

expected resonance frequency:

1 f0 = 2π

s

S ≈ 35 kHz 0 meff

(4)

Experimentally the resonance frequencies of the piezos (xed on one end and OC on the other end) are measured to be around 13kHz (see section 6.5). The measured values are approximately 2 times smaller than the theoretically expected ones. REASON?!?! The external cavity formed by the OC can be described as a Fabry-PerotResonator. The modes of such a resonator are spaced by blabla.

2.2 Internal & external cavity modes The laser frequency is mainly determined by the internal and external cavity modes as well as the transmission maximum of the IF. Here some important quantities: The IF maximum has a FWHM of ca 150GHz. The internal cavity has a length of 750µm. The modespacing therefore calculates (taking

nGaAs = 3.6

into account) to

∆νinternal ≈ 55GHz. ∆νinternal ≈ 1.7GHz,

The modespacing for the external cavity is

assuming a

cavity length of 8.5cm. So eventually, the laser should be emitting at the frequency that has the highest gain.

Figure 7 shows the spectral properties of all theses modes in

principal. The mode spacing, as well as the peak heights are not to scale, but more to get an idea how mode tuning works. The situation shown is kind of an ideal situation, the maximum of a internal and external cavity mode matches the maximum of the IF. In fact this perfect situation will almost never be realized in the lab, the modes will be slightly shifted agains each other. To tune the laser frequency one has now three options:

12

Figure 7: IF transmission mode + internal and external cavity modes determine the actual laser frequency.

1. Rotating the IF leads to a frequency shift of the transmission maximum. This can be used for coarse tuning. 2. To shift the internal cavity modes, one has to change the laser diode temperature or the laser diode current (what essentially also changes the temperature of the laser diode but only on a much faster timescale!) 3. By changing the piezo voltage, the OC mirror gets displaced, what essentially moves the cavity modes (it also changes the mode spacing, but that eect is negligible). The piezo voltage and the laser diode current can both be used for ne tuning and feedback in a laser lock scheme. The diode current can be used for very fast feedback (order of MHz) and the piezo usually for a few kHz. (more details see section 6.5) An idea how to get near to this optimal conguration is described in section 21.1.1. To get a feeling of how strong the dierent mode really are and what the frequency scales are, the internal and external cavity modes are plotted in Figure 8

13

Figure 8: Left picture: External cavity modes assuming R=1 for the back face of the diode: red curve: R=11%, blue curve: R=20%; Right picture: internal cavity modes, assuming R=1 for the back face and R=0.0003 for the AR coated face (datasheet). Lower picture: Both in one plot (external for R=11%). The plot range is the FWHM of the interference lter!

14

Figure 9:

Output power vs laser diode current for the laser setup with

ROC = 10.9%

3

(Laser 1) for dierent laser diode temperatures.

Output Power

The output power has been measured using a power meter. Figure 9 shows the temperature dependence of the output power of Laser 1. With lower temperatures the output power seems to increase. This could be an intrisical property of the diode. Other than that, it could be that due to the temperature change, the modes and the emitted frequency are changing. The modes at the egde of the FWHM of the IF (near a mode jump) have a lower gain and are therefore expected to have less output power. But this eect would average out, since with the current, also the wavelength gets tuned! However, the eect could be observed during the measurements, that near mode jumps the slope of the curve gets smaller. This can be nicely seen at high diode currents currents in the black curve, that the output power slope goes down every ~20mA (what corresponds to the current needed to scan over the free spectral range of the internal cavity - see section 5). Comparing the output powers of all the laser systems testet (Figure 10), one can see the following: The output power of laser 1 is higher than the output power of laser 2, even the OC reectivity is smaller. One explanation could be the fact that the feedback of laser 2 was maybe not optimal (see comment in Figure 10). Another explanation could be that with the R=8% mirror, the region of decreasing out-

15

OC reectivity 8% 11% 17% 22% bare diode

Ithreshold with IF 33mA±1mA* 26mA±1mA 27mA±1mA 25mA±1mA 65mA±1mA

Table 2: Threshold currents for all laser setups measured with the technique explained in section 19.2 (except the bare diode). (*maybe not optimized feedback due to not nicely collimated intracavity beam)

put power, when decreasing R is reached (when R gets too small, the properties of the bare diode dominate, what leads to lower output power when decreasing R). Surprisingly laser 3 (R=22%) has a higher output power than laser 4 (R=17%). This and the result described before, could indicate that in laser 4 (which uses the same setup as laser 2) is maybe something intrinsically worse than in the other setup. It could be just that the laser diode properties are dierent or that the optics are better in one setup (or less dirty). Also the laser diode in one setup (which provides the back mirror of the cavity) could have more of an angle to the z-axis than in the other setup, due to imperfections in the mechanical elements or during the gluing process of the diode. The threshold currents of laser 4 and 1 are also indicating, that the feedback in the two setups could be dierent (Table 2). However, with all lasers output powers of 40mW are possible below a diode current of 120mA, for laser 1, 3 and 4 even below 100mA, what's only half of the maximum diode current! The limitation on the diode is according to the spec. sheet an extracavity power of 100mW, but somehow there is no information given about the feedback of the external cavity, which determines the intracavity power (that's the one that could kill the diode!). Just for safety I would recommend the diode not to run at more than 100mW intracavity power, so depending on the OC reectivity one should set limitations to the current.

4

Beam prole

Using the Beam'R the beam prole of the four laser setups has been measured. The pictures are in the Appendix B. According to a gaussian t, all lasers are estimated more than 90%

TEM00

mode. The further away you measure, the better the beam quality gets. For example laser 4, the at a distance of 13cm (Figure 11) the prole looks really crappy, but at a distance of 125cm (Figure 12) the beam prole looks really nice! At short distances I expect eects like reections of apertures and the uo-

16

Figure 10: Output power for the 4 dierent laser systems, tested.

The laser

diode temperatures during the measurements were all around room temperature. *After unassembling laser 2, one could see that the beam after the rst collimation lens was not nicely collimated, thus the optical feedback could have not been optimal.

17

Figure 11: Laser 4 beam prole, left 13cm Laser 1

Laser 2

Laser 3

Laser 4

-2.2±0.3

-2.3±0.4

-2.1±0.3

-2.2±0.2

171±25 ≈ 43GHz

77

current tuning rangeinternal

≈ 44GHz

69±17 ≈ 40GHz

88±12 ≈ 43GHz

typ. current tuning rangeexternal

0.4-0.5GHz

0.4-0.5GHz

0.4-0.5GHz

0.4-0.5GHz

(dν/dI)internal [GHz/mA] (dν/dI)external [MHz/mA]

dν/dUpiezo

[MHz/V]

piezo tuning range

−53 ± 2

-41

-52

-42

>7.3GHz

>6GHz

>7.4GHz

>6GHz

Table 3: Tunabilities and sensitivities of all laser setups

rescence of the laser diode (that is still there and has a broad angle distribution) to mess the beam prole up. The beam coming out of the laser diode is usually elliptic, but the external cavity seems to clean the mode prole. The ratio of vertical to horizontal FWHM of the intensity is always measured to be

0.9 < FWHM < 1.1.

The FWHM does

not change signicantly, by changing the laser output power. Due to an astigmatic beam, it is not possible to collimate the output beam in both the horizontal and vertical direction.

It seems, that collimating it in

the vertical direction is somehow not nicely doable.

So the light was always

collimated in horizontal direction The collimation has been done using an IR card, which might not be the perfect tool to do that, due to saturation eects.

5

Wavelength tuning

As described in section 2.2, the ne tuning of the laser frequency is mainly done using the laser diode current and the piezo voltage. To nd out what the frequency sensitivities, as well as the mode hop free tuning ranges are for the piezo voltage and the diode current, the setup in Figure 13was used. The measurement results are shown in Table 3 Actually all measurements are limited by the 100MHz resolution of the wavemeter.

Especially for the current tunability within one external mode,

this limits pretty much the measurement accuracy! (the error bars are just the standard deviation taken for an average of 2 values!) Surprisingly the piezo mode hop free tuning range is bigger than the mode spacing of the external cavity modes. This is due to the fact that the internal cavity modes are very weak. The advantage of having such a big mode hop free tuning range is at the cost of stability: Imagine the laser starts at the optimal position shown Figure 7. Assume the external cavity mode gets shifted by more than the free spectral range without a mode jump. Then the neighboured mode is in this optimal position and should have the highest gain, but the laser is still

18

Figure 12: Laser 4 beam prole: upper picture 13cm distance, lower picture 125cm distance

19

Figure 13: optical setup used for most of the measurements

emitting in the other mode! One explaination could be, that tuning a running laser is dierent compared to tuning a not running laser. In the running laser there are already many photons in the cavity and if the frequency gets tuned adiabatically, the total gain is not only determined by the cavity modes, but also by how many photons are already in the cavity, using the population inversion for their frequency! One could test this instability by tuning more than the free spectral range, turning o the laser and turning it on again to see if it is lasing at the same frequency or at the one having the new optimal frequency. Unfortunately if the laser current gets turned down, the internal cavity modes get shifted what changes the hole situation and it is not clear how reversible this process is. What could be done is just blocking the beam in the laser with a piece of paper for a short time and therefore restarting the laser without chaning any of the relevant parameters.

6

Noise and linewidth

First a denition of important quantities is given, followed by an estimation how driver noise eects the linewidth.

Then several measurement results are

presented, to characterize the spectral properties of the dierent laser systems.

20

noise

typical time scale

typical frequency scale

temperature drifts

&1s

.1Hz

10ms-50µs

mechanical vibrations

100Hz-20kHz

electrical noise

up to a few MHz

hmhm Table 4: Typical frequency scales of noise

6.1 Denition: Linewidth, Drift and instantaneous linewidth The light emitted by a laser is not perfectly monochromatic, but has more a gaussian or lorentzian frequency distribution (or a mixture of both called

ν0

Voigt distribution) with a center frequency

and a FWHM of

∆ν ,

called the

linewidth of the laser. Common eects that contribute to this linewidth are the doppler eect (for lasers using a gas as gain medium), mechanical vibrations of the cavity (e.g. acoustic vibrations), temperature uctuations and noise of the controllers (current supply for laser diode and voltage supply for piezo). The time scales for this dierent eects can be seen in Table XX. Depending what the measurement time is compared to the time scale of the broadening eect, the contribution from this eect appears as linewidth or drift. Frequency uctuations slower than the measurement time will occur as a movement of the center frequency from measurement to measurement, what is called a

drift.

Frequency uctuations faster than the measurement time can't

be resolved as movement of the centerfrequency.

They lead to a distribution

of the center frequency during the measurement, what eectively broadens the line. The width of this distribution is called

linewidth.

So stating a result of a linewidth or drift measurement only makes sense, in combination with providing the measurement time. The linewidth that would be seen, having a measurement time innitesimally short (innite frequency) is called the

instantaneous linewidth.

For lasers, a minimum on the instantaneous linewidth is set by Heisenbergs uncertainty principle, but usually there are other limiting eects that broaden the instantaneous linewidth.

6.2 Contribution of driver noise to linewidth The piezo voltage driver has a noise level of

9.9mVpp

and

1.5mVRMS

[datasheet]

(no information about the frequency of the noise), tested without an external load connected. Adding a capacitive load, such as a piezo will decrease the noise, since the capacitance will create a low pass lter with the output resistance. So this numbers should be seen as an overestimated upper limit. The frequency sensitivity of the piezo is

dν/dU = −0.053 GHz/V according to the measurement

in 5. Therefore the overestimated expected frequency noise, caused by the noise of the piezo driver, would be

∆νRMS . 80kHz

21

respectively

∆νpp . 520kHz

.

The Current driver for the laser diode has a noise level of

< 1.5µARMS

(rip-

ple: 50/60 Hz & noise without ripple 10Hz..10MHz) [datasheet]. The frequency

dν/dI ≈ 170 MHz/mA ∆ν . 250kHz .

sensitivity is

and therefore the expected frequency noise

6.3 Noise spectrum In order to nd out at what frequencys the dominant noise contributions are, the noise spectrum has been measured. The trick is to turn uctuations in the laser frequency into intensity uctuations, using the saturation spectroscopy signal. This can be done by tuning the laser frequency to the edge of the crossover peak (red marked region in Figure 14). At this point the intensity is very sensitive to small movements in frequency, what can be measured using a photodiode (PDA36A, 17MHz bandwidth). By connecting the photodiode to a spectrum analyzer (HP 3585B, 20Hz-40MHz), the noise spectrum can be obtained. Since the lasers are free running, I checked once in a while that the frequency is not drifting away from the slope. After turning o the sweep for the sat spec signal, the function generator was unplugged from the piezo driver input, to avoid noise caused by the function generator (that gets multiplied by 15 at the input of the piezo driver). The noise spectra are pretty similar for all laser setups. An example is shown in Figure 15 and Figure 16. There are some noise peaks in the region around a kHz. This is the spectral range of sound waves and mechanical vibrations. Lowering the noise caused by mechanical vibrations could be done by setting the laser on a rubber pad, to decouple it vibrationally from the laser table. Although it's not really clear yet where this noise is coming from or over what channel it is brought to the cavity (base plate vibrations, cable vibrations, or just sound waves hitting the resonator at a mechanical resonance). In the region higher than 1.4kHz no measurable noise could be found (maybe limited by the noise oor of the spectrum analyzer) The noise peak at 1.4kHz does not change when the casing of the laser is opened, so its not a soundwave resonance property of the casing.

6.4 Beat note measurements One easy method to nd out what the linewidths of the lasers are, is to set up a beat note measurement.

6.4.1

Theory

The beams of two lasers, having the same polarization, are superimposed on a photodiode. The electric elds of both beams add and the photodiode makes an intensity measurement: with frequencys

ν1

and

ν2 ,

− → − → I ∼ |E1 + E2 |2 .

the intensity has

− → − → E1 and E2 as sinusoidal frequencys of ν1 − ν2 and ν1 + ν2 .

Assuming

The summed part is much to fast for any photodiode and averages out. If the

22

Figure 14: Saturation spectroscopy signal: on the edge of the cross over peak, the intensity is very sensitive to frequency uctuations (steepest slope).

two laser frequencys a very close to each other (on the order of a GHz or less), the dierence frequency can be measured using a fast photodiode (e.g. several GHz bandwidth) and a spectrum analyzer. The peak in the spectrum at ν1 − ν2 has a certain width given by ∆νbeat = ∆ν12 + ∆ν22 . Assuming that the linewidths of each√laser ∆ν1 and ∆ν2 are equal to ∆ν , the beat note width is given by ∆νbeat = 2·∆ν . This assumption

p

should be valid, since the lasers of each pair are identical.

6.4.2

Measurement

Several beat note measurements have been taken using the spectrum analyzer HP E4402B (9kHz-3GHz) and the photodiode ET 2030 FC (30kHz-1.2GHz). The photodiode was used without using a ber, to avoid back reections from the ber facet. But still the ber coupling input of the photodiode was used with the bare beams. For total sweep times of 50ms and 1s over a span of 15-16MHz, 20-30 single sweep measurement were made. Since the lasers are not locked, the beat note frequency moves around on the order of several hundreds of kHz (beat note frequency) over a few seconds. To compensate for that, all measurements have been recentered. To compensate for the fact that the jittering in some measurements broadens the line due to a movement in the scan direction and sometimes narrowing it due to a movement

23

10Hz - 2kHz:

10Hz - 50kHz:

Figure 15: Left: noise spectrum of laser 4, right: noiseoor without signal

Figure 16: Left: noise spectrum from 1Hz to 1MHz for laser 1. Right: noise oor without signal

24

Laser pair

low R

high R

low R

high R

sweep time, span

1s, 15MHz

1s, 16MHz

50ms, 15MHz

50ms, 16MHz

∆νGaussian [kHz] ∆νLorentzian [kHz] range for t |ν − ν0 |Gaussian range for t |ν − ν0 |Lorentzian sweeptime over ∆νGaussian

214 ± 27

193 ± 13

153 ± 7

139 ± 10

24

12

24

12

linewidth

linewidth

2MHz

≈ 10ms

≈ 0.5ms

>0.1

>2

contributing noise [kHz]

Table 5: Beat note measurement results.

opposite to the scan direction during the measurement, an average is made over 4-5 single measurements (average in the linear scale!). The choosing of the single sweep measurements is kind of arbitrary, one could only choose the narrowest ones in order to get better data. I tried to choose narrow ones as well as broader ones. I tried to not choose unsymmetric ones and ones that are obviosly messed up by mechanical or acoustic disturbances like hitting a table, slamming a door or Dylan talking and hitting the 1.4kHz resonance (see section 6.3) with his voice. Examples of the beat notes are shown in Figure 17. All other measurements look quite similar. The low frequency noise (around the peak) is due to technical frequency noise of the drivers and therefore has a gaussian distribution. The high frequency noise, is due to intrinsical limitations (for example the linewidth of the laser transition) and is thus given by a lorentzian distribution. Both regions has been tted, translating these distributions into logarithmic scales.

For the lorentzian t the boundary condition of matching the peaks

value was chosen. Linewidths measured for 1s, 50ms sweep time, as well as the intantaneous one (lorentzian t of the wings), are getting smaller, with higher OC reectivities. The gaussian linewidth dierence for the 1s and the 50ms sweep time indicates, that there is noise in the frequency region between 100Hz and 2kHz an addition in linewidth of 21kHz, respectively 14kHz is picked up. In this region the main two main noise peaks could be observed (see section 6.3) By not perfectly recentering the beat note measurements and taking the average, the linewidth could be broadened.

6.5 Piezo resonance frequency The resonance frequency

ν0 of the piezo is a very important quantity to know, es-

pecially for locking the laser frequency to an atomic transition. Locking means, that small uctuations of the laser frequency are corrected, using a very fast electrical loop circuit. The idea is, that whenever a small change in laser frequency is detected, the piezo voltage and/or the laser diode current get slightly changed in order to cor-

25

Figure 17: Upper picture: Beat note measurement: average over 5 single sweeps, 1s sweep time for laser 3&4; Lower picture: Beat note measurement: averaged over 5 single sweeps, 50ms sweep time for laser 3&4.

26

rect for the frequency movement. The higher the frequency of the uctuations, the higher the bandwidth of the lock circuit needs to be. The piezo is usually used for feedback in the frequency region up to a few kHz. This means, that the lock circuit can correct for laser frequency uctuations or drifts on a time scale down to several hundreds of

µs.

That region includes

temperature drifts and most of the accustic noise. Since the response of the piezo gets a phase shift in the region around the resonance frequency, the piezo would perform the reaction to uctiations at the wrong time. So one should cut the bandwidth of the piezo lock signal below

ν0 .

This can be done using a low pass lter and designing the -3dB threshold to be below

ν0 .

What's also important to know in this context is the width of the resonance

∆ν .

This quantity gives information about the region of non negligible phase

shift around the resonance frequency.

6.5.1

Measurement

The measurement is done using the same setup as in REFF, but instead of using the voltage driver for the piezo, the function generator is directly connected to the piezo. It drives it sinusoidally with an oset of 3V and an amplitude of 4V. The laser wavelength was set to the edge of the doppler broadened absorption line, so that a change of voltage translates to a small change in intensity. Then the intensity shows the same sinusoidally modulation as the voltage. The resonance frequency can be found by looking at the nonlinear amplitude response or second by looking at the phase dierence between the driving voltage and the photodiode signal. We've chosen the second possibility, since a drift in the laser wavelength would change the position on the edge of the doppler broadened absorption line and therefore the slope and thus the amplitude response. The phase shift however is independent of wavelength drifts. It is well described by the following formular:

 ϕ = where

ν

is the driving frequency,

−arctan ν0

2νν0 γ ν 2 − ν02

 (5)

is the resonance frequency and

γ

is a di-

mensionless damping factor. Driving below the resonance frequency leads to a phase shift above to

ϕ > 90°

and at the resonance frequency

ϕ = 90°.

ϕ < 90°, driving

Equation (5) is used

to t the experimental data (Figure 18), obtained for piezo 1 (the one in Laser 1 and 3) with the out-coupler R=22% glued on top. The damping ts to

γ = 0.025 ± 0.002

and the resonance frequency to

ν0 = 13.9kHz ± 0.1kHz. The width of the resonance (FWHM) is given by

∆ν ≈ 2 · γ · ν0 ≈ 0.7kHz

Similar measurements has been done for all used laser setups. Table 6 shows the measured resonance frequencys and the FWHM of the resonance.

27

Figure 18: Phase shift between driving and driven signal for piezo 1 with OC R=11%

28

Piezo

Outcoupler

Resonance Frequency [kHz]

FWHM [kHz]

1

R=11%

15.6 ± 0.1 13.0 ± 0.5 13.9 ± 0.1 13.0 ± 0.1

0.6

2

R=8%

1

R=22%

2

R=17%

0.7 0.5

Table 6: Piezo resonance frequencys and width of resonance. The OC R=22% is accidentally glued a little bit o center to piezo 1.

The OC R=22% was accidentally glued o center to piezo 1, what might explain the decreasing of the resonance frequency compared to the OC R=11% on top. But this also shows, that gluing a OC o center has just a small eect on the resonance frequency. For piezo 2 the resonance frequency stays the same for both OC's.

6.5.2

Consequences for laser locking

For the piezo to expand and contract, a certain current is needed. This current depends on the driving frequency (sinusoidal driving assumed) as follows:

ˆ

max current needed/provided by driver (one should do this calculation just to see if the bandwidth is limited by the maximum current provided by the piezo driver!!)

Part II

Instructions for building the ECDL This part explains step by step, how to build an external cavity diode laser with an interference lter as the wavelength selective element. It is assumed that all mechanical parts are already machined and the reader has an idea of how they should t together. Details about the theory behind the laser can be read in Part I

29

7

Schematic of the laser

Figure 19: Optical setup of the laser

8

Needed items

8.1 Drivers & controllers - Current driver for laser diode (ref: Thorlabs ITC502) - Temperature controller for laser diode (ref: Thorlabs ITC502) - Temperature controller for base (ref: Thorlabs TED200C) - Voltage driver for piezo (ref: Thorlabs MDT694A - Single Axis)

8.2 Mechanical parts - Base plate - Cavity main part - Mounts for cat's eye lenses (brass) - Mount for piezo + out-coupler

30

- Mount for laser diode (copper) - Copper tube for heat conduction - Mount for collimation lens - Mount for SUB-D & BNC connectors For Pictures see Appendix A

8.3 Optics & electronics - Collimation lens (ref: Thorlabs C230TME-B) - Cat eye lens 1 (ref: Thorlabs 52280-B) - Cat eye lens 2 (ref: Thorlabs 52220-B) - Laser diode socket (ref: Thorlabs S8060) - Piezo ring actuator (ref: Piezomechanik HPSt500/10-5/5 length 9mm) - Peltier for base (ref: Digi-key 102-1682-ND) - Peltier for laser diode (ref: TeTech CH-41-1.0-0.8) - Out-coupler (ref: CVI Melles Griot PR1-767-08-0512) - Heat sink (ref: Digi-key 504222b00000) - Interference Filter (ref: Research electrooptics) - Laser diode (ref: Eagleyard EYP-RWE-0790-04000-0750-SOT01-0000 ARcoated) - 2 thermistors (ref: Newark 30C7947)

8.4 Miscellaneous - Thermal conductive glue (ref: Loctite 384) - Thermal grease (ref: ) - BNC connectors with soldering cups - BNC cables (RG58) - Acetone & tissues (for cleaning purposes) - SUB-D15 & SUB-D9 connectors both male & female with soldering cups - Wires - Misc screw drivers and hex keys

31

Figure 20: Ultrasonic bath: Water in the outside bowl and acetone in the inside bowl.

- Water level - Shrink tubes - Misc nylon washers

9

Cleaning of mechanical parts

After machining, the mechanical parts are often dirty, oily and chips are sticking in small corners and threads. So all parts have to be cleaned before assembling the laser: 9.1 Prepare an ultrasonic bath and ll it with water. 9.2 Put some of the parts into a beaker and ll it with acetone until every part is covered. Place the beaker into the water lled bowl of the bath (see Figure 20) Turn it on and wait for 10-15 minutes.

Note: The dirt has the tendency to fall down to the bottom of the beaker, so in order to clean all faces of the parts, turn them around after each run and do the cleaning again (letting them stand diagonal in the beaker might safe some runs). 9.3 Dry the parts after cleaning and store them on a clean surface, for example in aluminum foil. Do these steps until every part is cleaned.

10

Wiring - Electric Connections

For electronic connection from the laser to the voltage/current supplies and the temperature controllers the following conguration is used: On the back of the laser are two spaces for BNC connectors (laser diode current, piezo voltage) and one for a SUB-D15 connector (for thermal management) (see Figure 10)

32

Figure 21: Back part of the laser

Purpose

Designation

Pin # (SUB-D15 f )

Pin # (SUB-D9 m)

LD - Thermistor (+)

Laser Diode

2

2

10

3

LD - Peltier (+)

1

4

LD - Peltier (-) Ground

9

5

LD - Thermistor (-) Ground

Base - Thermistor (+)

Base

Base - Thermistor (-) Ground Base - Peltier (+) Base - Peltier (-) Ground

5

2

13

3

4

4

12

5

Table 7: Pin allocations for thermal management cable

Get some SUB-D9 (female/male) and SUB-D15 (female/male) connectors with soldering cups on the back ready. In principle all critical soldering connections, meaning connections that could possibly touch each other or that carry high voltage, should be covered with shrink tubes.

10.1 Thermal management wiring The SUB-D15 connector is used for the thermal management, meaning the peltiers and thermistors for the base and the laser diode. For the pin allocation see Table 10.1. Since the peltier and thermistor for the base are connected to a dierent temperature controller than the ones for the laser diode, you have to split the cable into two parts as shown in Figure 10.1.

The double end of this cable gets connected to the delivered cables of the temperature controllers via the SUB-D9 male connectors. Table 10.1 shows the

33

Figure 22: Cable for thermal management: SUB-D15 female on one side and 2x SUB-D9 male on the other side

Figure 23: BNC to SUB-D9 female

pin allocations of the two SUB-D9 male connectors compared with the ones of the SUB-D15 female on the other side of the cable. So make sure the cables go the right pins! Twist the cables before soldering to reduce noise picked up by electromagnetic elds in the lab. Cover the connectors of your self made cable with SUB-D shells.

10.2 Wiring for laser diode current Use a BNC cable for the laser diode current due to good shielding, since the emitted frequency of the external cavity diode laser is very sensitive to small changes of the laser diode current. For the piezo also use a BNC cable. The output connector of the laser diode current driver is a SUB-D9 male, so one has to transform it to a BNC type connector by using a SUB-D9 female connector (see 10.1). Solder the connections according to Table 10.3. Also solder a small wire from Pin #1 to Pin #5 to short circuit the Interlock of the Thorlabs ITC502 current driver. (check the Pin # in the instructions manaul if you use another current driver). Cover it with a shielded SUB-D shell. Glue the BNC connector to one side of the shell, to avoid stress on the wires.

34

Connector

Purpose

Designation

Pin BNC

Pin # SUB-D9

BNC for Current

Laser Diode

Current (+)

inner

8

Current (-) Ground

shield

3

BNC

Piezo

red (+)

inner

black (-) Ground

shield

Table 8: Pin allocations for laser diode current and piezo voltage

10.3 Wiring for the piezo The piezo gets connected to its voltage supply by a BNC cable. So just install a female BNC connector at the back plate for now.

11

Assembling: out-coupler to piezo

Choose a clean, at and nicely horizontal surface for the gluing. Get some slow glue ready (e.g. Epage 11, regular epoxy). Don't use 5 minute epoxy, since it degases more than slow epoxy and degassing could damage the coating on the mirror. 11.1 Place the out-coupler mirror (OC) on a few layers of optical paper (or fold one paper a few times without touching the relevant section) and make sure the partial reecting coated side points up (see the arrows on the side of the mirror). Make sure the up pointing side is very clean and blow all dust particles away with the hand dust blower since after the gluing it's very hard to clean this side of the mirror! 11.2 Place the piezo on top of the mirror in the right orientation (the cables of the piezo should be near the OC mirror) and center it. 11.3 Place a weight very gently on top of the piezo and check by eye if it is still centered on the out-coupler (see Figure 24). Choose the weight with regard to not damage the out-coupler or the piezo, but heavy enough to keep a nice, parallel contact between the two. Make sure your weight puts pressure uniformly on the piezo! 11.4 Glue the piezo with 3 glue points to the mirror. Make the glue points very small, since the glue expands on the mirror surface while curing and one doesn't want the glue to expand to the center of your mirror!! In Figure 25 on the left the expanded glue can be seen (dark spots). One can see that one glue point has expanded to the inner circle of the piezo, so in this case the limit is reached! Figure 25 on the right shows an better example.

Hint 1:Use a thin wire and dip the tip into the glue, then dip it to the point you want to set your glue point, until both orthogonal faces are connected by the glue. This should be a reasonable measure of how much

35

Figure 24: Assembling: Piezo to out-coupler with weight on top

Figure 25: Glue expansion (dark spots): On the left picture the limit is reached! On the right picture glue has been set well dosed.

glue you need for one point.

Hint 2:Mix the two components of the glue, but wait for approx. 1 hour before setting the 3 glue points.

During that hour the glue gets more

viscose. Therefore the glue expands less while curing and it is easier to make very small glue points. If a dierent glue is used, do some test runs on a glass sample to see the expansion properties of the glue in dependence of preparation time before setting glue points. 11.5 Check again if everything is centered and let it cure over night.

12

Assembling: piezo+out-coupler to cat's eye mount

After nishing the gluing in chapter 11 check by eye if everything is nice and parallel and still in good condition. If anything went wrong remove the glue by

36

putting some acetone drops on the glue and try to scratch it away with a spiky item. Then do chapter 11 again. Sometimes also soapy water is a good solvent for glue. Also check if the glue has expanded to the inside of the piezo ring. If so, contact an experienced person in your lab to discuss further instructions. If everything is ne do the following steps: 12.1 Get the mount for the piezo ready and cleaned (especially the surface the piezo sits on). Also make sure the mirror is still clean and not damaged. Get a mechanical help construction for gluing ready (see Figure 28). 12.2 First get the wires of the piezo through the hole inside the mount while keeping the piezo with the OC on it in your hands. This helps the piezo to keep its position later on since the wires are very springy. Don't touch the mirror on the faces! Now get the piezo down to it's position. Try to play around with the wires to make the piezo sit in position by itself. 12.3 Prepare a few layers of optical paper (2 or 3 layers, not to thick!) cut it, so that it ts into the mount hole and lay it down on the OC mirror. Place the mechnanical help construction softly upside down on the OC mirror and center it. Never touch the mirror directly with the metal, make sure optical paper is always between the two. 12.4 Turn the whole (help construction+piezo+out-coupler+mount) upside down, while keeping little pressure on the help construction to let it all stick together in its position. 12.5 Now that it's upside down, you can see the back of the piezo from the other side of the mount.

Center the mounts hole to the piezo hole and

check if the stick of the mechanical help construction is also still centered to the hole of the mount! If everything is centered, glue the piezo from the back with 3 small glue points to the mount. (see Figure 26). 12.6 Place a weight on top of the mount, to make sure everything stays nice and parallel! Place it gently to not change the centeredness of your assembling! (see Figure 27) Wait one night to cure.

13

Soldering and testing:

peltiers, thermistors

and laser diode socket Before assembling the whole laser, the peltier and thermistor should be tested, rst to make sure they work at all, second to nd out the correct orientation of the peltier and third to nd out if your self made cable works and all soldered connections are ne. Therefore proceed as follows:

37

Figure 26: Three glue points on the back of the piezo

Figure 27: Piezo+out-coupler to cat's eye mount assembling: Help construction on the bottom and weight on top

38

Figure 28: Mechanical help construction to keep the piezo+out-coupler in position while gluing

13.1 Soldering 13.1.1 Screw a male SUB-D15 connector with soldering cups to the back plate of the laser. 13.1.2 Cut some shrink tubes to a reasonable length and put them over the wires of the peltier for the base. Solder the peltier to the right pins of the SUBD15 (see Table 10.1).

Check the datasheet for the polarity (should be:

red = +, black = - ) 13.1.3 Do step 13.1.2 with the peltier for the laser diode. 13.1.4 Since the leads of the thermistors are to short, add a thin piece of wire to each lead. The best way to connect these two is to twist them around each other and solder the connection. 13.1.5 Solder the added wires to the SUB-D15 (both the base & laser diode thermistor). Make sure to put the shrink tubes for every solder joint on the wires before, but do not heat them up yet. 13.1.6 Take a female BNC connector and solder two small wires to it, one to the shield pin and one to the inner pin. The length should be so, that they can reach the back of the laser diode which sits on the copper housing later on, better a few cm longer!

Connect these two wires to the laser

diode socket with the conguration shown in Figure 30.

Twist the two

wires around each other to avoid noise pickup!

13.2 Testing Do the testing of the thermal management without any use of thermal grease or glue!

39

Figure 29: Peltiers and thermistors soldered

Figure 30: View from the back of the laser diode socket

40

13.2.1 Screw the back part of the laser to the base plate and get the peltier for the base in its position. 13.2.2 Place the cavity main part on top (no need to screw it) and get the thermistor for the base in its hole. 13.2.3 Connect your self made cable to the SUB-D15 and the part for the base to one of the temperature controllers, using the original output cable. 13.2.4 To test if the thermistor works, simply touch it with your hands and watch if the temperature controller changes the Tact value. To test the orientation of the peltier, set Tset to a value near room temperature, turn the peltier control on and watch if the temperature converges towards this point. If it diverges, turn the peltier around and do the same test again. 13.2.5 Repeat the same procedure with the peltier and thermistor for the laser diode. If possible do this testing without having the diode glued to the copper mount, to not damage it. 13.2.6 Test the laser diode socket and its wiring with a simple LED using the current controller. Twisted pair wiring is at least for the thermistor wires recommended!

14

Assembling: cavity to base plate

Now it's time to get the cavity main part on top of the base plate.

14.1 Cavity, peltier and base plate 14.1 Clean the surfaces the peltier gets pressed to and the surfaces of the peltier as well. 14.2 Cover the down pointing surface of the peltier with heat sink compound/heat conductive paste (for example GC electronics Type Z9 Heat Sink Compound) with a very thin layer (use a razor blade or a sharp knife). Press the peltier softly and symmetrically from the top into its place on the base plate.

It should stick to the base plate quite tightly, so that lifting the

peltier up again should be hard. If not, do it again and use little more heat sink compound. 14.3 Cover the second surface of the peltier also with a thin layer of heat sink compound and press the cavity main part on top. Use the M4 nylon screws to tighten it down on the peltier. Don't screw it hard, since peltier surfaces can be cracked easily. Try to apply the force symmetrically by screwing every nylon screw one by one just a little bit. Use a level to make sure the force is symmetrically and the output beam will be aligned straight. To get a feeling for a reasonable amount of force use a small screw driver

41

Figure 31: Cavity on top of base plate (peltier in between)

and screw without putting pressure on top of the screw driver and wait until it jumps out of the notch by itself. Then give it a little bit more till you feel a resistance. Don't squeeze the wires of the peltier! 14.4 The excess of the compound on the side of the peltier should be minimal!

14.2 Thermistor for base 14.1 Clean the hole for the thermistor and put some heat conductive glue into it. 14.2 Try to ll the hole without bubbles, using a sti thin wire (or solid solder). 14.3 Cover the thermistor with activator and stick it into the hole. 14.4 Give it a few hours to cure.

15

Assembling: laser diode to copper housing

The laser diode gets glued into the copper housing as shown in Figure 32. While assembling keep the following rules in mind:

ˆ

Never touch the laser diodes face

ˆ

Laser diodes are very sensitive to electrostatics.

Always wear the hand

wrist thing to ground yourself !

ˆ

Never wear rubber gloves while touching the laser diode, since rubber gloves can carry electrostatics, even if you wear the hand wrist thing.

For the assembling you can twist the pins of the laser diode around each other to reduce the inuence of electrostatics. Test carefully if the laser diode ts into its mount and take it out again.

42

Figure 32: Laser diode glued to copper mount

15.1 Prepare some tape (which sticks well to metal) and cut a small round piece out of it with the size, to cover the laser diodes front head.

Also

cut a smaller round piece of lens cleaning tissue and put it onto the tape with the size to fully cover the diodes front face (see Figure 33).

Tape

it carefully to the diodes front head, so that the lens tissue part covers the hole diode face and the tape sits tight at the edges of the head. (see Figure 34) 15.2 Take some thermal conductive glue (for example Loctite384) and cover the upper part of the hole in the copper mount with the glue. Don't use to much glue, since to much excess is a mess and could get onto the diodes face! 15.3 Cover the laser diode with the activator and stick it slowly into the hole. Watch the excess and clean eventually before pushing the diode completely inside. Keep in mind that the orientation of the laser diode determines the orientation of the polarization. The polarization axis is important, because the interference lter has dierent reectivitys for dierent orientations of polarization according to Fresnel's formulas, when it sits on an angle later on. The electric eld is polarized in the plane of the small notch on the side. (see Figure 32 for the right orientation) 15.4 Clean the excessed glue from all surfaces and remove the tape from the diodes face. 15.5 Give it one night to cure.

43

Figure 33: Cover for laser diode face (round is better see Figure 34)

Figure 34: Laser diode with face cover

44

16

Assembling:

laser diode+copper housing to

cavity

16.1 Preparation: Cat's eye mount to cavity This assembling will take place on the optical table to avoid electrostatics damaging the diode. ALWAYS USE THE HANDWRIST THING! First glue the copper ring to the heat sink, using thermal conductive glue. Just put a thin layer of glue on one side of the ring, put the activator to the heat sink and glue them together centered. Put a weight on top and let it cure a few hours (or better one night). Then take the mount for the cat's eye conguration and use a tweezer to get the wires of the piezo out of the hole again, because you can't attach it to the cavity with the leads hanging out. Try not to touch the mirror. Push the part into place and use three M2 screws to screw it tight into position. After that, use the tweezers to get the piezo leads through the hole again. Solder them to the BNC connector (or better solder some wires to the BNC connector and then the leads to this wires). Cover all connections with shrink tubes.

16.2 Assembling To get an overview of how the laser will look like after the following steps see Figure 35.

This might help understanding how the parts t together.

This

assembling is one of the most dicult and delicate ones in this instructions manual. 16.2.1 First clean the surfaces the peltier gets attached to and the peltier's itself. 16.2.2 Make sure you have tested the orientation of the peltier! 16.2.3 Now put thermal grease on the face of peltier which gets attached to the laser diode copper housing. Make a very thin layer, using a razor blade, similar to step 14.1.2.

Try to avoid thermal grease spreading into the

region between the two peltier surfaces.

If you want to, you can try to

cover this part with tape, but you should remove it before assembling! 16.2.4 Also cover the surface of the copper ring with a thin layer of thermal grease. Get the copper ring in place and stick the socket through the hole, without the copper rings surface touching anything! 16.2.5 Untwist the laser diode pins and cut them to a reasonable length according to the size of the socket. 16.2.6 Push the peltier to the copper housing in its place and make sure the surface contact is nice (maybe take it o again just to see how much of the surface has had contact). The contact should be good if the peltier sticks to the housing and it's hard to get it o again. Next connect the socket to laser diode pins.

45

Figure 35: After assembling. Layers on the screw axis from left to right: Screwhead, nylon shoulder washer, copper housing, nylon washer, cavity, air, heatsink, hexnut, hexnut.

16.2.7 Get the two M3 screws (25mm) and the nylon washers in place (shoulder washer and spacer) on the copper housing and screw the copper housing very tight to the cavity! The peltier should stick to copper housing by itself while screwing (if not, start again).

While screwing push the heat sink

in place, meaning its holes over the screws. After tightening the copper housing to the cavity really nice, push the copper ring against the peltier, it should stick! Take the nuts and screw them from the back tightly, but do not crack the peltier. Unscrew the back connector part to have more space to tighten the nuts. Use a second set of nuts to keep the rst nuts position x. 16.2.8 Glue the thermistor for the laser diode into the hole using thermal epoxy. Do it the same way you did with the base thermistor. 16.2.9 Screw the laser to the table and congure the PID control for for base and laser diode (see manual of temperature controller).

After that keep the

temperature controllers always running. Test if the laser diode is working by turning on the current (make sure cathode ground is selected and the adjustment button is turned on) and looking for uorescence with an IR-card 16.2.10 If you see red light coming of the diode, make a jump and shout: yeaaaaaah!.

17

Optical setup for testing and optimization

In the next chapters the lenses will be aligned and optimized to make the laser as ecient, powerfull and stable as possible. Then some saturation spectroscopy on Potassium vapor will be done to test if every part of the laser is working ne as well as to nd the best angle for the interference lter and to optimize the emission frequency to the Potassium line.

46

All these steps can be done with the optical setup shown in Figure 36 without changing, moving or realigning components.

There may be hundreds of

other setups to perform the optimization but this uses a minimum of optical components and it's very easy to set it up, but feel free to create an own setup ;) It might be convenient to align the setup as recently as going through the following sections, just because a collimated beam is needed to align it.

Figure 36: Optical setup for optimization

18

Collimation lens

The light coming out of the laser diode is divergent and needs to be collimated. Always turn the laser o, while changing or adding parts to the cavity. Assemble the collimation lens as follows: 18.1 Wrap some teon tape around the thread of the lens to have a better adjustability due to more screwing resistance (see Figure 37 ). Screw the lens into the round brass mount and push the brass mount into the hole in the aluminum mount. 18.2 Get the aluminum mount over the diode copper housing and use 4 M3 screws (25mm) and metal washers to attach it to the cavity.

You may

have to cut one edge of the washers at for the lower ones in order to make them t.

47

18.3 The fact that the holes for the screws are bigger than the screws themselves gives you a certain freedom in x-y-alignment for the collimation lens. You denetely need this freedom, because you want to align the beam, coming from the collimation lens, nicely through the middle of the piezo ring. Is the aligning freedom not enough, modify the aluminum mount in the machine shop. 18.4 Turn the laser diode current on to a value of ca 100mA. Try to coarsely align the aluminum collimation lens mount, that the beam is going through the piezo and tighten the screws. 18.5 Now push/pull the movable round brass mount to coarsely collimate the beam. Keep in mind, the bigger the observation distance, the easier it is to collimate the beam. Bring the brass mount to a position, that the beam is slightly not collimated in a way that you can correct it by screwing the lens a tiny bit out. Glue the brass mount in this position with 3 glue points of 5 minute epoxy. 18.6 Then collimate the beam nicely by using the screwing degree of freedom of the lens. If the beam has astigmatism, collimate it just in the horizontal direction. After some alignment you may feel that the screwing resistance gets less and less, so change the teon tape once you have the feeling it's to sensitive to align the lens. 18.7 Rise the laser diode current up to 150mA so that you can see the edges of the elliptic beam prole on the edges of the piezo ring. Losen the screws of the mount a bit and try to center the beam as nice as possible to the center of the piezo by symmetrisizing the reections on the edges. Thighten the screws and keep in mind that the beam moves slightly while tightening them. 18.8 Once again test if the beam is nicely collimated. If not collimate it and check if the beam is still centered to the piezos center.

19

Cat's eye lenses

19.1 Preparation In this section you don't need the vapor cell, so take it out. Put the two cat's eye lenses into their mounts. If they don't t, use some very ne sand paper to make the hole for the lens slightly bigger. Make sure to also press the sand paper against the outer wall deep inside the hole while rubbing it , to take o material homogeneously! Get the lens in place above the hole. If it doesn't fall in by itself, lay down some layers of optical paper on the lens and try to push it softly with your nger. Once it's in, make sure by eye it has gone all the way to the bottom of the mount. Then use 3 glue points of

48

Figure 37: Teon tape around collimation lens thread

slow glue on the side of the lens (less degassing and therefore less damaging of the coating of the lenses) to x its position in the mount.

19.2 Observing the threshold The goal in this subsection is to actually modulate the current around the laser threshold and make the threshold visible on the osci. The reason is, that the threshold current is a measure for the quality of your external cavity feedback, the better your feedback is the lower the threshold. Therefore you can optimize the lens positions by minimizing the threshold current. Turn the laser o and screw the lenses into the cat's eye mount. Turn the current on again to 50-100mA and align the distances between the lenses and the out-coupler coarsely, use the second lens to collimate the out coming beam. Now set the laser diode current to a value around 40mA and modulate the laser diode current with a triangle signal of 1V peak to peak and 10-100Hz using a function generator. Simply connect the output of the function generator to the modulation input on the current driver and the sync output to the oscilloscope. Also connect the photodiode to the oscilloscope. So try to get a signal similar to the one shown in Figure 2.

By changing

the laser diode current (oset) you can move the position of the ellbow in the curve. To measure the actual threshold, just change the current oset, so that the ellbow is in the middle of the modulation period (think about what you see on the picture and nd out where the middle of the modulation period is) and turn o the modulation. The value of the laser diode current shown on the driver now is the threshold current!

19.3 Optimizing the lens positions By turning the rst cat's eye lens you can probably see the threshold moving towards smaller oder bigger values. Try to optimize the lens positions by min-

49

Figure 38: Threshold observation

imizing the threshold current. You basically have two degrees of freedom, the position of the lens in the cat's eye and the collimation lens rigth after the laser diode. Use the following steps to optimize both positions (nding a minimum in a 2D search space)*: 19.3.1 Minimize the threshold by turning the rst lens. 19.3.2 Slightly change the position of the collimation lens (the one right after the diode) by screwing it in. 19.3.3 Re-optimize the threshold current by turning the cats eye lens. Sometimes the laser beam moves when you are turning the lenses and you could therefore lose the signal on the photodiode. Try to compensate this beam movement by realigning the mirrors to always have the full picture as shown in Figure 2 19.3.4 There are two possible cases now: 1. the threshold current is bigger than before 2. the threshold current is smaller than before If case 1 is observed, do steps 19.3.5, 19.3.6 and 19.3.7. If case 2 is observed, iterate the steps 19.3.2 and 19.3.3 until the threshold currents gets bigger again and get the collimation lens back into the position of the lowest threshold. Now the feedback of the external cavity is optimized. You can also try to maybe nd an even smaller minimum by screwing the collimation lens more and more out again (speaking going into the other direction), but its not very likely to nd more than one minimum. Go on with step 19.3.8 19.3.5 Screw the collimation lens slightly out. 19.3.6 Re-optimize the threshold current by turning the cats eye lens. 19.3.7 Iterate steps 19.3.5 and 19.3.6 until the threshold current gets bigger. Then get the collimation lens back to the position where the minimum

50

Figure 39: Assembling of the interference lter to the barrel

has been observed and re-optimize the cats eye lens. Now the feedback is optimized! *At the end I just collimated the beam with the rst collimation lens (and no other lens in the laser) very nicely in the horizontal axis and then put the cat's eye lens and the outer collimation lens in and optimize only the cat's eye lens. The reason is, that the FWHM of the interference lter gets eectively broadened when the beam is not nicely collimated inside the cavity. Second reason is, that the threshold doesn't change signicantly change when you change the collimation lens. In other words, the maximum in the 2D search space is very at in the direction of beam collimation.

20

Interference lter

The interference lter is the wavelength selective element in the laser. Assemble it according to the following steps: 20.1 Use tweezers to place the lter into the notch in the barrel. 20.2 Fix the lters position with an helping construction as shown in Figure 39. Make sure it is nicely aligned vertically without an angle. 20.3 Take slow glue (low degasing) to glue the lter with two very small glue points on the side to the barrel. 20.4 Let it cure one night. 20.5 Wrap some teon tape near the bottom around the barrel. This makes the angle alignment easier, due to more resistance, once the lter is sitting in the round notch in the cavity main part.

51

21

Saturation spectroscopy

Get the interference lter into the notch in the cavity and bring the laser diode current to 50-100mA. This should be enough to observe the saturation spectroscopy signal.

21.1 Finding the Potassium line - Fluorescence 21.1.1

Filter angle - centering the transmission maximum

With the angle of the interference lter one can coarsely tune the wavelength. The following steps can be very hard, since you may need a very good feeling in you ngers. Make sure the teon tape provides enough turning resistance, to be able to rotate the lter barrel by very small angles. With too much resistance the barrel jumps from angle to angle, what's also bad. Rotate the lter and watch the wavelength changing on the wavemeter. The goal frequency for the potassium line is ca 391015GHz. Try to rotate the lter, so that the 391015GHz is in the middle between the two mode jumps of the internal cavity. Since the free spectral range of the interal cavity is about 40GHz, mode jumps should occur at 391035GHz and at 390995GHz. vary the temperature on the laser diode to nd these mode jumps.

Simply

Now the

lter transmission maximum should be centered right at the potassium line. After adding the interference lter to the cavity, the threshold current gets less. But usually you can optimize the threshold slightly again. Just modulate the current around the threshold again and turn the cat's eye lens to optimize it. The idea with the symmetric mode jumps holds for centering the lter transmission maximum to any frequency.

21.1.2

Fluorescence

Now that the frequency is coarsely aligned, the next step uses the laser diode temperature or the laser diode current to do some ner tuning. Watch the vapor cell with an infrared camera and tune the frequency around the 391015GHz using the current and/or the temperature till you can see uorescence on the IR camera.

The uorescence should occure right where the

beam is going through the cell. Write down all settings!

21.2 Observing saturation spectroscopy signal Modulate the piezo voltage with a frequency of 50-200Hz and an amplitude of approximately 10V peak to peak (10Vpp should be on the piezo, think about gain of modulation input of the driver!) sinusoidally or triangular. Connect the sync output of the function generator to the osci and trigger the photodiode signal to it. Use AC coupling for the photodiode signal. Now one should see a drop in intensity on the AC photodiode signal at some point of the modulation period and little peaks next to it (see Figure 40). If not, use

52

Figure 40: Saturation spectroscopy signal, with a heated K cell (ca 50°C)

the piezo voltage oset knob to scan left and right of the spectrum for it. Once it has been found, center it in the spectrum and have fun identiing the little peaks you see ;) To improve the visibility of the peaks in the saturation spectroscopy signal, you can heat up the vapor cell to 40°C or 50°C. Using the saturation spectroscopy signal, one can play around with dierent parameters (laser diode current, piezo voltage, temperature) to get a feeling for the mode hop and wavelength tuning behaviour of the laser.

21.3 Optimizing external and internal cavity modes 22 ˆ

What to improve 3 screws for copper item instead of 2 could improve vibration stability of the diode copper housing, even if it is not clear if this is an issue... just for safety. This would also reduce the possibilty of a tilting of the diode relative to the optical axis.

ˆ

Better adjustability for the collimation lens in z direction, using a screwable brass mount, instead of a movable.

ˆ

Better adjustability for the collimation lens in x-y-directions, without loosing the rigidity of the mount in z direction, relative to the laser diode! I'm not sure yet how to do that. It is not mandatory, more a nice to have but it would denitely improve the hole design and adjustability.

ˆ

(More safety in apertures, maybe a piezo with thinner walls, but apertures make it more likely to be in TEM00)

ˆ

Make sure a very good job in the machine shop is done, setting important screw holes and dimensions which dene the optical axis!!

53

ˆ

The assembling of the heat sink, copper housing etc (section 16) is very tough!! Easier screwing of the hex nuts would be nice.

ˆ

Heat sink vertical.

ˆ

Adapt the cavity main part hole to the shape of the peltier and adapt the copper cylinder for heat conduction to that shape.

ˆ

Use thinner wires for thermistors and especially laser diode, to reduce inuence of cable movement to cavity.

ˆ

(Hole for the piezo wires in the cat's eye mount slightly bigger & set it at a nice position)

ˆ

Use rubber pad for vibration damping and foam inside the setup when the cover is on, to reduce soundwave resonator eects.

ˆ

23

Filter @ 767nm

Costs for one laser

54

55 Digikey inventory inventory inventory

nylon washer

misc wires and connectors screws glues

teon tape, thermal grease

Description

two needed

two needed

(better 767nm), FWHM 0.3nm

teon tape, thermal grease

5min epoxy, slow glue, thermal conductive glue

M4 nylon, M5, M3 25mm, M2,

SUB-D15 & SUB-D9 connectors, shells, wires

λcenter,0° = 781nm

see lab book for drawings, appendix for photos

λc ≈770nm

Table 9: List of needed items for one laser including prices and overall price to build one laser

≈ $7000 ≈ $6500

. $30

$2,00

$3,00

unknown

≈ $2000

$604,55

$0,61 Eagleyard, AR coated,

$1,60

10kOhm, isolated leads, two needed

Tution for the education: $50000, Buying all the parts: $7000, seeing it work for the rst time: priceless ;)

with 15% Thorlabs discount

$26,40 $205,00

thermal Resistance@75°@natural: 6.4°C/W

R=8%, DIA=0.500", Thickness=0.125", 767nm

22.5mm x 17.5mm, ID=9.5mm (hole)

$16,10

$342,00

40mm x 40mm, 50W

$968,00

OD 10mm, ID 5mm, L= 9mm, 0-500V, stroke 7µm

$1.860,00

$712,00

$4,21

$84,00

$75,00

$87,00

Price

12W

I=200mA, TEC 16W

150V output, small signal bwdth 10kHz (no load)

for 9mm diode

f=11.0, NA=0.26

f=18.4, NA=0.15

f=4.51mm, NA=0.55

Overall price (without IF, discount for Thorlabs products not applied yet)

Seastrom 5607-225 Seastrom 5610-10-93

nylon shoulder washer

Machine shop Research electrooptics

mechanical parts

EYP-RWE-0790-04000-0750-SOT01-0000

laser diode

interference lter

Newark 41K9376

CVI Melles Griot PR1-767-8-0512

out-coupler mirror Digi-key 504222b00000

TE-Tec CH-41-1.0-0.8

peltier for laser diode

heat sink

Digi-key 102-1682-ND

peltier for base

thermistor

Piezomechanik HPSt500/10-5/5 9mm

Thorlabs MDT694A - Single Axis

voltage support Piezo

piezo ring

Thorlabs S8060

laser diode socket Thorlabs ITC 502

Thorlabs 352220-B

lens L2

Thorlabs TED200C

Thorlabs 352280-B

lens L1

temperature control base

Thorlabs C230TME-B

current & temperatur control LD

Company & Partnumber

Item

collimation lens

References [Clairon] [Rasel] [1]

A

Photos of Mechanical Parts

Figure 41: Copper housing for the laser diode; the thermistor hole should better be set on a small angle!

Figure 42: Mount for collimation lens mount

56

Figure 43: cat's eye mount

Figure 44: Base plate

57

Figure 45: Laser cover

58

Figure 46: Cavity main part

Figure 47: Back part for electronic connections

59

Figure 48:

Mounts for the collimation lens (middle) and the two cat's eye

lenses

Figure 49: Heat sink

B

Beam prole pictures

60

Figure 50: Laser 1 beam prole, 60cm distance

Figure 51: Laser 2 beam prole, 30cm distance

61

Figure 52: Laser 3 beam prole, 30cm distance

Figure 53: Laser 3 beam prole, 110cm distance

62

Figure 54: Laser 4 beam prole, left 13cm

Figure 55: Laser 4 beam prole, 125cm distance

63