Minutes of the Project coordinators meeting on April 1, 2019 (DRAFT 1)

1. Part A. [Agenda item #1]

Presentation by Dr. Alexandre Mermillod-Blondin, On-chip Laser-written Photonic
Circuits for Classical and Quantum Applications.

Dr. Mermillod-Blondin described the basics of laser-direct writing of photonic
circuits, in particular as applies to reconfigurable photonic chips. In fused
silica, three regimes of modification are possible, depending on the pulse
duration and power: 
- uniform increase of the refractive index (low pulse energies); 
- formations of self-ordered nanoplanes (this regime is sensitive to the
  polarization of the field); and 
- cracking (high energies, long pulses).

There are multiple origins of the material modification, namely:
- point defects
- reorganization of SiO2 structure from the 6/7-membered rings in the normal
  silica to smaller rings after irradiation, leading to density increase
- induction or relaxation of stress in the material (e.g. waveguides are
  usually made in a pre-stressed material)

It is possible to write waveguides in the variety of shapes, including
straight, curved, etc.

The limitations of direct laser writing include:
- limits on the radius of curvature: it is difficult to make radii <30mm
- high induced stress, leading to poor mechanical integrity of the waveguides.
- the writing is mostly within the bulk of the material. It is hard to write
  waveguides close to the surface of the material.

A key application of direct laser writing is in reconfigurable photonic chips.
For example, gold electrodes can be added. Heating of a chip with electrical
current changes the propagation constant gamma. The difficulty with this
approach is the long equilibration time, on the order of 15 ms.

Another key application is in surface waveguides. Such waveguides can be
written by using few-cycle pulses and grazing incidence. The typical losses in
surface waveguides and below 0.1 dB/mm. The waveguides are sensitive to the
surface state. For example, adding a drop of oil can increase attenuation to
over 2.5 dB/mm. This sensitivity can be increased by adding the right material
on the surface.

An important development direction is towards building surface waveguides with
programmable losses, for example by coating the surface with an electro-optical
polymer. The possible applications of such programmable-loss waveguides
include:
- Combining the surface and volume waveguides, to explore non-Hermitian
  photonic systems.
- Sensors: limited coupling between surface and bulk waveguides, leading to
  formation of exceptional points. Adding a single molecule to the surface then
  pushes the system beyond the exceptional point, leading to large changes in
  signal intensity.

There is a unique combination of skills at MBI for developing field, with
experiment, measurement, and modeling all available in-house.

Prof. M. Ivanov pointed out that the theoretical methods presented to describe
the interaction between the waveguides are identical to the ones used to
describe interference stabilization and auto-ionization, like for example what
he did on his own thesis, and now Nicola Meyer has applied to N2. He suggests
to play with the phase shift between the waveguides to increase the contrast.

Dr. A. M-B asks if this means independent control on phases and voltage for
each waveguide, and Prof. M. Ivanov answers yes.

Dr. O. Kornilov points out that investigation of exceptional points is currently 
a hot topic, and asks how this research is related to switching of modes?

Dr. A. M-B answers that it is th same thing, that there are many papers, but he
does not like to call it mode-switching.

It was asked how do the written surface waveguides compare to spliced fibers?
Is it possible to combine them with fibers? Is it possible to reduce the losses
within the waveguide?

Dr. A. M-B answers that the minimum loss is approximately 0.05 dB/mm, likely
due to inhomogeneities. The losses can generally be reduced by backing, and are
mostly scattering rather than absorption.

It was asked whether surface waveguides can be also written in fibers?

Dr. A. M-B replies that yes, this is possible.

It was asked that such work is already possible in lithium niobate. What is the
advantage of using silica?

Dr. A. M-B replies the it would depend on the application. Generally, silica
has better contrast.

It was asked that one can produce similar structures lithographically in SiN.
Why use direct writing?

Dr. A. M-B replies that direct-writing system has 3D capability, which is hard
to achieve with SiN. Furthermore, direct writing can also be used in silicon.

2. Part B. 

[Agenda item #2] Minutes approval.

There were no objections to the minutes, and were approved.

[Agenda item #3] Brief presentation by Dr. C.P. Schulz about Good Scientific
practice

Discussion:

The question about the read access rights for each publication was raised

The present agreement is: ombudsperson, archivist, contacts 1 and 2 (optional)
and directors.

The GSP guide of the DFG recommended to store data 7 years in a durable form.

A discussion about use of data and agreement with the institute followed, but
it was clarified that the institute can decide who has read access or not.

It was also discussed the definition of primary data: is it all data, or just
the data required for the final published figures?  It is pointed out that
future agencies might require to make public also primary data, but it is yet
to be defined the rules of what to publish and how.

Dr. Patchkovskii points out that there is no other requirement so far but to
store the data, no public access.  He also asks the question if the current
system and infrastructure would be compatible with this?

Prof. Elsaesser : The difference is that when you store all primary data, even
what does not enter publication should be stored. He also points out that
everyone should also be aware of the differences of DFG requirements vs
Leibniz:
* DFG requires acceptance of these rules for application to DFG funding
* Leibniz are just recommendations
Neither require to store all primary data

Prof. Ivanov says that storing primary data would be too much, and Dr.
Patchkovskii says that it could be done but it has a huge overhead.

Prof. Eisebitt says that for many experiments is impossible. Dr. Morales points
out that the purpose of the GSP is to be able to reproduce published results 

Dr. Nibbering: CD-DVD approach does that.
Dr. Schulz says that we should also update our own text and rules on the website.

Prof. Elsaesser brings the discussion back to the who has access question to
download the files from archive. So far the ombudsman is clear, but project
leaders should also be added, in case the person leaves the institute, we need
someone with access. 

Dr. Patchkovskii says that the directors are added automatically. 

Prof. Elsaesser argues that it could lead to many people asking for files.
Prof. Elsaesser also reminds that the policy is that all data should be here @
MBI, all data generated at the institute should be stored at a MBI computer.
Legally all data is property of MBI or FVB.

Dr. Patchkovskii then points out that, for example, he cannot act in the name
of MBI of FVB when accessing data.

Prof. Elsaesser argues that towards inside the institute yes, of course towards
the outside, all directors represent the institute.

The defaults should be ombudsperson, archival GSP (Librarian), project
leader(s), primary and secondary contact, directors.

Dr. Patchkovskii points out that some projects/publication have more than one
project coordinator.

Dr. Kruel ask if this change (adding coordinators) should be implemented?

The vote was unanimously in favour of adding the coordinators.

Dr. Kruel then demonstrated draft GUI design for GSP. Mentions that the
handle-id cannot be changed, project coordinator will be added to list, there
is no coupling for endnote for metadata (authors, doi, etc...)

Prof. Elsaesser asks if is there a check for required files, e.g. the pdf of
the paper?

Dr. Kruel answers that no but it can be implemented.

Prof. Elsaesser says that at least it should be there: the original manuscript
+ figures + original files with data, numbering according to figures.

There is a vote to decide between two possible implementation of the check of
the required files:
- Use a standard naming scheme for each publication: the software makes sure
  that these files exist in the directory (e.g. paper.pdf, readme.txt).
- Give a list of filenames to the librarian that correspond to these files.

Option one (Standard naming) wins 9 votes to 7

[Agenda item #4] Other Items:

Dr. Rouzée asks what is expected from us for evaluation and by then.

Prof. Eisebitt answers that there is the online schedule, and that the
time-line will be updated.

For the moment, important appointments are: 
- Apr 9th: project coordinators should present project strategy - 15 min each. 
- Apr 17th publication documenting
- Apr 18: posters to be checked by directors and Misha, template will be sent 
  around
- May 2nd Test print + standing in front of posters to test.

Prof. Elsaesser reminds that the number of posters per project is 3, one for
overview (why project is done, scientific motivation is important, key results,
direction of project). The remaining two posters are for the highlights.  As
the evaluation panel does not know the institute so well, it is important to
show clearly why each project is interesting.

This format is different from that we use for the Advisory Board, members of
which are more familiar with the institute.

Prof. Vrakking points out that the overview poster should be consistent with
the information included in the evaluation report. The highlight posters can
provide additional information.

Prof. Ivanov points out that there are 5 theory sub-groups, and that 2 posters
only would be tricky in terms of visibility of the individual sub-groups.

Prof. Eisebitt also reminds that at the end of April all labs should be labs
clean, all in good shape.  As all panel members visit labs, time is limited, so
people should rehearse what to say on each lab. Labs to be visited will be
defined by directors.

Dr. Patchkovskii asks if there are any plans to show theory facilities?

Prof. Elsaesser says if its possible, if the panel members ask.

Prof. Eisebitt finishes reminding that there will be an institute wide meeting
to tell everyone about it.

Minutes prepared by F. Morales and S. Patchkovskii on June 4-28, 2019