1. Progress on Optical Rotational Cooling of SiO+
9/3/ :25 PM
Progress on Optical Rotational Cooling of SiO+
Patrick Stollenwerk / Yen-Wei Lin / Brian Odom
Molecular Ion Trapping Northwestern University
© 2007 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.
The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

2. Outline
Motivation: Ground state preparation and coherent control of a molecule
Challenges and Advantages of SiO+
Rotational Cooling
Fluorescence imaging as state readout
Give a brief motivation for why we want to do this. Some characteristics of our molecule of choice and then how we plan to rotationally cool and use fluorescence imaging to determine if we have done so.

3. Ba+ Ion Trap Coulomb crystal, 0.6mm x 0.2mm (x 0.2mm).
Ion string, 0.8mm long.
In our group it is routine to trap atoms, which you can see fluorescing here with your naked eye, in a linear paul trap. Once they are in the trap we can keep them there essentially for forever, interrogating the same set of ions over and over again.

4. Ba+ Ion Trap Coulomb crystal, 0.6mm x 0.2mm (x 0.2mm).
Ion string, 0.8mm long.
We are capable of cooling them down to mK temperatures where they begin to form ordered structures such as this coulombs crystal

5. Ba+ Ion Trap Coulomb crystal, 0.6mm x 0.2mm (x 0.2mm).
Ion string, 0.8mm long.
We are further able to manipulate trapping parameters such that we can get them to align into a string and study interesting things this way.

6. Ba+ Ion Trap Coulomb crystal, 0.6mm x 0.2mm (x 0.2mm).
Ion string, 0.8mm long.
We are even capable of loading single ions and trap them indefinitely. Hold them A LONG TIME. We would even keep them for so long that we would name them.

7. Ba+ Ion Trap Coulomb crystal, 0.6mm x 0.2mm (x 0.2mm).
Ion string, 0.8mm long.

8. The same, but more! I.E. Image and coherently control molecules.
What Do We Want?
The same, but more! I.E. Image and coherently control molecules.
Why? e- p+
Ultracold chemistry
NJP 11, (2009)
Want to have the same control over molecules. This is much more difficult because of extra DOF. So why subject ourselves to these extra complications? Well because in every new challenge lies a new opportunity! And molecules have some very promising opportunities that atoms do not have. For example a time variation of p/e could be directly seen with a change in the frequency of a rotational or vibrational transition. Control over molecules would also be highly prized for their ability to study ultracold chemistry and how internal states affect chemistry.
Sensitive probe to time variation of fundamental constants
Quantum information processing

9. The SiO+ Datasheet B2Σ+ X2Σ+ 28Si: 92%; 16O: 99.8%.
Zero nuclear spin: no hyperfine structure.
(Nearly) Diagonal Frank-Condon Factors in B-X
160 (5000) photons without (with) broadband vibrational repumping
~43 GHz spacing between lowest rotational states
Small predicted dissociation cross section from B-state of order cm2 (difficult!)
SiO+ potential curves
B2Σ+
X2Σ+
Now that we have our motivation let’s look at our candidate. One that SiO doesn’t have that reduces its complexity relative to other molecules is hyperfine structure. However, the most important feature to notice is the highly diagonal FCFs. These are important because it allows us to address rotational transitions at optical wavelengths without exciting vibrations. Another nice feature is that the rotational spacing is easily accessible with today’s microwave technology. A minor downside is the small predicted cross-section for REMPD from B state. However, this is not so bad as we are more interested in a non-destructive state readout rather than a destructive one.

10. First Challenge: Loading
First attempt: ablation loading. It works for Ba+!
Laser ablation produces Na+, Si+, SiOH+ etc., but not much SiO+.
Ref: J. Phys. Chem. A 113, (2009).
Mass Spec of ION SPECIES Produced from Ablation
SiO+, m=44
First challenge was putting it in our trap! EMPHASIZE IONs PRODUCED. SiO is a solid like Ba so we figure ablating it like barium should also work. Good idea bad result. As you can see the ions produced from ablating SiO are mostly not SiO. Not very good for loading a pure crystal.
Oven doesn’t really work. ~11 eV ionization energy.
SiO in test vacuum chamber

11. A Solution! 2+1 REMPI
Ignore the ions produced by ablation and ionize the neutrals!
Ion signal yield vs. wavelength
Mass spec. in ion trap
HX(0,0)
HX(1,1)
HX(2,2)
SiO+
Ba+
Counts
Our solution was to ignore the ions produced from ablation and focus on selectively ionizing the SiO neutrals using REMPI. We were able to identify SiO in the REMPI spectrum and we see SiO mass in our trap.
Mass (amu)

12. Translationally Cold (mK) SiO+
Typically load 500+ Ba+ and 100+ SiO+.
Lighter mass more tightly confined -> dark core formation
Ba+ sympathetically cools SiO+
No short-range collision -> internal state remains hot
Dark 137Ba+ ions
Dark core: SiO+
Bright 138Ba+ ions
EXPLAIN DARK BA LOCATION. Great we can load it. So we typically load barium first which will then sympathetically cool SiO+. And because SiO+ is lighter it is more tightly confined by the RF trap and will therefore form a dark core which you can see here. Unfortunately coulomb repulsion prevents short-range collisions preventing cooling of internal states. So we’re left with cold motion and room temperature internal population which brings us to our next challenge.

13. Next Challenge: A Thermal Population
B2Σ+->X2Σ+(v=0,v’’=0)
R-branch
300 K
Relative Population
P-branch
1000 K
Addressing a thermal population. In SiO+ roughly 40 rotational states are occupied at room temperature. So this is a problem because we have to somehow take this distribution and push it all down into the ground states. This is where diagonal FCFs can help us. If we had enough resources and patience you might imagine it would go something like this…. ‘’

14. Rotational Cooling B X . . . . . . 385nm Rotational Cooling Laser(s?)
40 (+)
39 (-)
38 (+)
37 (-)
. . .
2 (+)
1 (-)
0 (+)
385nm
. . .
N’’=
40 (+)
39 (-)
38 (+)
37 (-)
2 (+)
1 (-)
0 (+) X
We drive all the p=branch transitions and, eventually, because parity is conserved throughout the cooling process we pump everything down into the lowest two ground states. 40 states, means 40 lasers => terrible. So we have to be more clever
Rotational Cooling Laser(s?)
Population build up in lowest two parity states

15. Pulse Shaping for Rotational Cooling
MaiTai FS laser source (Spectra-Physics) used for pulse shaping
SiO+ P/R branches are separated so a razor blade acts as a sufficient mask
Technique already demonstrated on AlH+
Basic idea is to separate frequency components using grating and filter out the ones we don’t want. p/r branch =>razor blade. Mention previous talk.
Nature Communications 5, 4783 (2014)

16. Spectral Filtering
wavelength
This is the result of our pulse shaping. The cut off resolution is more challenging because rotational spacing. Think we are a little better than this. 90 to 10 extinction Ba transition as calibration. After pumping down into ground state we need a way of reading out the state
Drive P-branch transitions until everything is pumped into rotational ground state (~1-10 ms). Cooling rate currently limited by laser intensity.

17. State Detection: Quasi-Cycling Transition
N J P
Photon budget
160 without any repumping
~5000 with vibrational repumping v=1, 2, 3..
B(v=0)
P11(1)
385 nm
Semi-closed
τ =~350 μs
τ =11 μs
In our case, due to selection rules, the P1 transition provides us a quasi-cycling transition. Where we can on average expect 160 spontaneous emissions before an off diagonal decay into an excited vibrational state. If we are capable of repumping all the vibrational decays this number goes up to 5,000. Sufficient for single ion imaging. Of course, in general, repumping all these vibrational states is easier said than done.
P12(1)
τ =70 ns A
N’’ J’’ P’’
X(v’’>0)
X(v’’=0)
8 MHz

18. Vibrational Repump
R-branches
P-branches
Our solution to this is in much the same vein as our rotational cooling solution. All the vibrational cooling transitions are close together and well separated (~1000cm-1) from other vibrational transitions. So in this case we can use a broadband repump without needing to do any pulse shaping. So, with our rotational cooling laser and vibrational repump laser prepared we have already begun to characterize our imaging system
Same idea as rotational cooling! SHG of Spectra-Physics Tsunami FS laser used as broadband source
All other vibrational transitions separated by >103 cm-1!

19. Photon Budget 0.1% Total Detection Efficiency (using Ba+ measurement)
With our source production established and all of our lasers ready we have characterized our CCD background in preparation for looking for a fluorescence signal. IMAGE A SINGLE MOLECULE FOR THE FIRST TIME IN FREE SPACE WITHOUT RELAXATION ASSIST. Before we try CW fluorescence we will attempt a much less precise pulsed fluorescence experiment first. As you can see, based on our signal estimates we hope to be getting a fluorescence signal very soon.

20. Overview Reliable loading method established Lasers:
FS broadband laser pulse shaping for rotational cooling cutoff independently measured
FS broadband laser vibrational repump nm ready
CW is prepared for fluorescence detection
Imaging background has been characterized

21. Outlook Pulsed fluorescence detection
Obtain first single molecular ion image in free space with CW laser
Begin microwave control of N=0 and N=1 rotational states
I should emphasize that this will be the first image of a single molecule in free space without any sort of assisted relaxation. Reiterate microwave easy.

22. Acknowledgments Our Group! Brian Odom PI Research Fellows
Matt Dietrich
Zeke Tung
Grad Students
Mark Kokish
Yen-Wei Lin
Chris Seck
Ming-Feng Tu
Past members
Jason Nguyen
Joan Marler
Chien-Yu Lien
Vaishnavi Rajagopal
David Tabor
Funding
Agencies:
Undergrad
Eugene Wu