17.01.2013 Views

Martin Teichmann Atomes de lithium-6 ultra froids dans la ... - TEL

Martin Teichmann Atomes de lithium-6 ultra froids dans la ... - TEL

Martin Teichmann Atomes de lithium-6 ultra froids dans la ... - TEL

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

3.7 MOT imaging<br />

3.7. MOT IMAGING<br />

When optimizing the MOT, it is of crucial importance to know the<br />

number and temperature of the trapped atoms. The old MOT imaging<br />

system had long stopped working, mostly due to the fact that the<br />

frame grabber card to read the camera did not fit into new computers.<br />

Therefore we <strong>de</strong>ci<strong>de</strong>d to install a completely new system.<br />

As for the camera, we <strong>de</strong>ci<strong>de</strong>d to buy a cheap and especially small<br />

camera, as we do not need precision measurements for the characterization<br />

of the MOT. We chose the uEye UI-1410-M from IDS, an eight<br />

bit CMOS camera, which is triggerable via a TTL signal and can be<br />

connected to a computer using the USB port, which at the same time<br />

is its power supply, limiting the number of cables on the experimental<br />

table. There was a problem with that camera. A g<strong>la</strong>ss sli<strong>de</strong>, meant to<br />

protect the camera, created strong fringes on the camera, such that the<br />

images were <strong>de</strong>stroyed by the interference fringes. Unfortunately, while<br />

breaking this g<strong>la</strong>ss sli<strong>de</strong> we <strong>de</strong>stroyed some pixels on the camera, but<br />

there were enough left over to work with the camera.<br />

As with the main imaging system, we use absorption imaging, since<br />

this gives an absolute number of atoms. Assuming we correctly installed<br />

the imaging lenses, each pixel of the camera will “see” a column of<br />

the imaged cloud. If the cross section of this column is A, and the<br />

scattering cross section of a photon with an atom is σ, a photon traveling<br />

along this column has a probability of σ/A to scatter with an atom.<br />

Consi<strong>de</strong>ring many photons, there will be 1 − σ/A of the inci<strong>de</strong>nt light<br />

intensity transmitted. For N atoms Iout/Iin = (1−σ/A) N will be left, calling<br />

Iin and Iout the inci<strong>de</strong>nt and transmitted light intensities, respectively. In<br />

or<strong>de</strong>r to know the number of atoms, we take the logarithm, leaving<br />

us with N = ln(Iout/Iin)/ ln(1 − σ/A) ≈ − ln(Iout/Iin)A/σ. Note that σ/A<br />

is normally sufficiently small that the <strong>la</strong>st approximation is very good.<br />

The term − ln(Iout/Iin) is called the optical <strong>de</strong>nsity. The advantage of<br />

absorption imaging is clearly visible: the calcu<strong>la</strong>ted number of atoms<br />

only <strong>de</strong>pend on the ratio between two intensities, which can be easily<br />

<strong>de</strong>termined by taking two images, one with, one without the atoms to be<br />

imaged, and we take the ratio of the two. Therefore there is no need to<br />

calibrate the calcu<strong>la</strong>ted number of atoms. The formu<strong>la</strong> for the scattering<br />

cross section is<br />

σ = 3c2λ2 2π (1 + 4(∆/Γ)2 ) −1<br />

(3.4)<br />

with λ = 671 nm the wavelength of the transition, Γ = 5,9 MHz its<br />

line width, c the corresponding Clebsch-Gordon coefficient, and ∆ the<br />

67

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!