Thread by @markmccaughrean

With the huge #JWST sunshield now successfully pulled out (I can’t believe I’m actually typing that after so many sleepless nights 🤯), the next step is to tension it into its proper shape.

That’s crucial too, so *our* tension is by no means over yet 😨

1/

https://twitter.com/markmccaughrean/status/1477208228593344514

As my long thread yesterday described, the sunshield plays the key role in establishing a temperature difference of ~300°C between the sun & space-facing sides of the observatory.

Only when the cold side reaches 40K or -233°C do we have the infrared performance we desire.

2/

https://twitter.com/markmccaughrean/status/1476836183103754243

So far though, the five wafer-thin metallised Kapton layers are in a relatively floppy state & touching each other. That means they can conduct heat & thus the cold side can’t, well, get very cold.

Thus the layers have to be separated & held apart from each other.

3/

Not only do they have to be held apart, they need to be splayed like the fingers on your hand spread wide.

That allows the heat bouncing between each progressively colder layer to escape into space.

This is called a V-groove radiator & @esa’s Planck mission had one.

4/

Planck was launched together with @esa’s Herschel mission in 2009, also on an Ariane 5 like #JWST. They were both infrared observatories too, albeit working at much longer wavelengths than #JWST, Planck to study the cosmic microwave background & Herschel the dusty Universe.

5/

Both were “small” enough (Herschel’s main mirror was 3.5 metres in diameter 🙂) to be launched without being folded up & both used passive cooling behind shields like the V-groove radiator on Planck to get their main optics cold out at L2 like #JWST.

6/

But when it came to their scientific instruments measuring the far-infrared light, both Herschel & Planck used liquid helium to get to the extremely cold temperatures needed to work at very long wavelengths. And that ran out after ~4 years, effectively ending those missions.

7/

#JWST uses passive cooling behind the sunshield to get its optics down to 40K / -233°C & to get three of its scientific instruments (NIRCam, NIRSpec, & FGS/NIRISS) to their operating temperatures. The longer-wavelength MIRI needs an extra boost from a cryocooler to reach 7K.

8/

So, no exhaustible liquid cryogens involved in #JWST – even MIRI’s cryocooler is closed loop, a bit like a household refrigerator, only at far colder temperatures.

That mean a longer lifetime for them, limited by other factors on the observatory. Which means more science.

9/

Why do the detectors need to be cold?

They use semi-conductor material to sense infrared photons, mercury-cadmium telluride (HgCdTe) in the near-IR instruments & arsenic-doped silicon (Si:As) in MIRI.

Like the silicon in your phone camera, but with a narrower band gap.

10/

The narrower the band gap in the material, the lower the energy an incoming photon requires to excite an electron from the valence band into the conduction band, after which it can be moved around & measured by the detector’s electronics.

(Sorry, getting a bit technical 😬)

11/

As IR photons have relatively low energies, you need narrow band gap materials to sense them. But that makes them temperature-sensitive. If they’re too warm, then electrons can have enough energy to randomly jump from the valence into the conduction band without a photon.

12/

We call this “dark current”, a signal from the detector itself, nothing to do with light. And if you get too much of that, it’s blinded, saturated or over-exposed in the minimum exposure time of the camera. Which means you can’t see the light photons you’re really after.

13/

In semi-conductors like silicon able to detect visible light, the band gap is wide enough that even at room temperature, the dark current is quite low. Although astronomers often cool their silicon CCDs & CMOS devices to reduce it further & allow long exposures.

14/

But in HgCdTe & Si:As as used by #JWST’s detectors, the IR photons have low energy, the band gap is narrow, & the dark current is higher.

So they need to be cooled to very low temperatures to avoid being saturated & allow long exposures of outer space.

15/

Which is why deploying the sunshield & tensioned is key: without that, the telescope won’t get cold & neither will the instruments.

If the telescope doesn’t get cold, it’ll glow brightly at the wavelengths where we’re trying to detect the light of faint stars & galaxies.

16/

And if the instruments don’t get cold, they’ll be blinded by their own dark current & useless.

So, while the overnight deployment of the sunshield is brilliant news, now all of the cables & motors & pulleys need to work to tension it & separate the layers.

Hang tight 🤞

17/

Then we still need to unfold the aft radiator, & the primary & secondary mirrors, before we have a deployed observatory.

And then the long cooling, optical alignment, & instrument commissioning begins – there’s another five & a half months of this to go 😱

18/18

Coda: there’s a subtlety to HgCdTe as a semiconductor. By changing the ratio of Hg to Cd, you can change the band gap & thus wavelength cut-off & dark current susceptibility.

NIRCam has both short & long wavelength cutoff HgCdTe detectors & that will become important soon 😉