Aller au menu principal Aller au contenu principal


Structural Study of biological macromolecules using crystallography

PROXIMA-1 (PX1) is one of two beamlines at SOLEIL for measurements in macromolecular crystallography (together with PROXIMA-2A). PX1, operational since March 2008, delivers an intense, nearly parallel and tunable x-ray beam for measurements at high resolution or from large unit cell dimension crystals. 

Equiped with a very large surface area detector (PILATUS 6M) and coupled to the latest generation three-circle Chi geometry goniostat, advanced data collection strategies are routine at the beamline for complex experimental phasing. High-throughput data collection takes advantage of a robust sample changer robot of the CATS family. Mostly targeted towards cryogenically cooled samples, the robot can also handle in situ sample plates, mostly for crystallisation plate screening.


PhD Student
CHAVAS Leonard
Beamline Manager
Beamline Scientist
LUBLIN Victoria
Post Doctoral Student
FOOS Nicolas
RICHET Nicolas
Post-Doctoral fellow at the Institute of Pharmacology and Structural Biology
+33 (0)5 61 17 54 03
student thesis
scientist of the beamline

Technical data

Energy range

Between 6,5 KeV and 15 KeV

Energy Resolution

2E-4 (Si 111)


U20 In-vacuum undulator

Flux and first optical element

White beam – depends on undulator settings


Kirkpatrick-Baez pair of bi-morph mirrors plus channel cut cryogenically cooled monochromator crystal

Sample Environment

3 circle κ goniostat (10 μm sphere of confusion)
Sample changer is currently unavailable.Oxford Cryosystems cryostream
Si drift diode energy dispersive detector plus MCA for fluorescence measurements

Beam size at sample

In general 40 x 20 µm2

Flux on sample

> 2.0 e+12 Phot/s/0.02%bw for 500 mA stored current.





Scientific opportunities

High resolution macromolecular
High resolution diffraction measurements (< 0.8 Å).
Large unit cell MX

Diffraction measurements from crystals with large unit cell dimensions (complexes, viruses).

Phasing by MAD and SAD

MAD / SAD phasing from most heavy atoms of interest for MX (Se, Fe, Hg, Pt, Br etc). Access to long wavelengths for S-SAD phasing.

Applied Research High throughput measurements – automated data collection, phasing, high resolution studies.




The source of the light line is a U20 inverter, containing 100 periods of 20 mm. This inverter was chosen to obtain the highest possible energies, without loss of brightness.


A transfocator can be defined as an optical equipment consisting in positioning lenses on the passage of a beam to modify its geometric properties. In our case, to overcome the chromatic aberration of Beryllium, this equipment must have motors to insert or remove the lenses and a waterproof enclosure to vacuum it. The lenses are positioned in copper stacks that differ in the number of lenses they have. Thus the first stack (closest to the source) has 4 lenses and corresponds to the focusing configuration for 6keV. The stack number 2 has 5 lenses and allows to cover an energy of 7keV. The stack number 3 has 6 lenses for an energy of 8keV. Finally, stack number 4 contains 7 lenses for an energy of 9keV. Then, it is enough to make different combinations of stacks to cover the highest energies until all the stacks are inserted for 15keV.



A monochromator is a device used in optics to select as narrow a range of wavelengths as possible from a light beam of wider wavelength range. On the PX1 light line we have a channel cut Monochromator made up of a silicon Si crystal (111), selecting a power range from 6keV to 15 keV with a very fast energy change time.

Mirror KB

A Kirkpatrick-Baez mirror, or KB mirror for short, focuses beams of X-rays by reflecting them at grazing incidence off a curved surface, usually coated with a layer of a heavy metal. X-rays can be focused by compound refractive lenses, these also reduce the intensity of the beam and are therefore undesirable. KB mirrors can focus beams to small spot sizes with minimal loss of intensity. Typically they are used in pairs - one to focus horizontally and one for vertical focus. When the horizontal and vertical focuses coincide, the X-ray beam is focused to a point.


The goniometer was created by the company SMARACT. It allows the analysis of crystal samples. The goniometer is composed of 5 axes. The Omega axis is the main one and can turn 360 °. It is the same for the Phi axis which is located at the head of the goniometer. The Chi axis, meanwhile, allows a vertical movement from 0 to 50 °. Each axis has a sphere of confusion. Their experimental measurements are: 1 micron inferior for the Omega axis 3 micron inferior for the Phi axis 2 micron inferior for the Chi axis


CATS sample feeding robot, with plate screening option. The robot uses the standard UniPUCK with 16 crystals in one puck. The DEWAR consists of 3 slots for UniPUCK.


Until September 2018, the detector was a PILATUS 6M (Dectris Ltd.). Today, we are equipped with the Eiger 16M detector (Dectris Ltd.).

Thanks to the latter, it is possible to work up to 120 Hz.

The master file contains experimental parameters → masterfilename.h5

But also experimental data → filename_data_numerodufichier.h5

Before the experiment

Equipment Available on the Beamline : 

  • Roentec SDD detector for measuring XANES spectra.
  • Oxford Cryosystemes Cryostream  700 series for sample cooling.
  • Various Dewars for manipulation and transfer of samples.
  • Tools and Uni-pucks for preparation of robot samples.
  • Users may borrow tools for sample mounting (magnetic rods, loops, vials etc). These must be returned to beamline staff after the experiment. 
  • PILATUS 6M detector has a maximum of 25 Hz.


Length of loops  (IMPORTANT) :

The goniometer is designed to accept SPINE standard loops. For details, check this link - .

We can accept loops between +-2mm compared to the SPINE standard 

Data backup :

Transfer by sftp is not supported on the line (neither for data import, nor for data export).

The range of sample translation on the goniometer allows us to mount loops up to +/-2.0 mm compared to the SPINE standard. Note that the loop length, as defined by the SPINE standard, includes the dimension from the magnetic base to the sample. Very long crystals in long nylon loops may exceed these dimensions and consequently be impossible to centre - consequently this is VERY IMPORTANT.

For robot users :

A CATS robot system is now installed on the beamline, and is operational. Operating instructions can be found .  The robot uses the UniPUCK standard with 16 crystals in a single puck. Tools for mounting crystals in UniPUCKs are available on the beamline, and kits for transfering crystals to the synchrotron in UniPUCKs can be loaned to beamline users. For details, please contact the beamline staff. It is best to bring crystals pre-mounted in UniPUCKS or on canes for mounting on site. 
A short video explaining mounting procedures :

A further video showing the use of a dry shipper and sample recovery :

Vial pucks are available at the beamline for transfer of samples from canes. This can only be used for SPINE standard pins AND vials.

IMPORTANT : In order to minimise the risk of problems or failures of the robot, we apply the following conditions to usage :

  • a) User must minimise the number of times they load pucks into the Dewar. In practise this means filling 3 pucks (48 samples) in the morning.  Collecting / testing data from 3 full pucks can last anywhere between 4 and 8 hours.
  • b) Passage from robot usage to manual usage will only be permitted once per day, in either direction and in the presence of the local contact.

The robot will also have the capacity for screening crystals in crystallisation plates.  For details of plates supported, please contact the beamline staff.


Backup : 

Data cannot be transferred to or from the beamline by sftp, though individual data sets (not complete days of data collection!)  can be transferred via the SunSET (for details please contact the beamline staff).

Backups can be made on external USB disk via a LINUX station. We have 5 disks that can be temporarily loaned to users for backup. Type "synchrosync" on any LINUX station for data processing - the script prompts for the mount point of your USB disk, the name of any subdirectory and the directory from which you wish to back up. The script works by launching Rsync regularly, and will continue to incrementally back up new or modified files appearing in this directory tree. With the PILATUS detector, it is not uncommon to collect up to 500 Gb per day - come prepared with sufficient space.

You can arrange to connect your laptop to the experimental area network if you have declared this in advance using your user account on  SunSET.


Local Contacts / Starting and finishing times :

A standard experiment will be given 3 shifts of beamtime which start at 08:00 am and end at 07:00 am the following day. Between 07:00 am and 08:00 am every day the beamline team perform routine checks to verify the performance and alignment of the beamline.

Unless he or she has warned you in advance, your local contact will be available on the beamline at 08:00 am to help you start your experiment and explain how the beamline functions.  Your local contact will be available, either on the beamline or by telephone, between 08:00  and 23:00 the day of your experiment.

For technical or scientific questions outside these working hours, you must refer to the beamline documentation, the SOS "help desk server" or the experimental area co-ordinator (telephone 97 97). If none of these approaches solves your problem, you should stop measuring and go to bed UNLESS your problem could engender damage to beamline equipment - this is the only case in which you may ring your local contact outside the above working hours.



The monochromator crystal has now been changed, resulting in an improved bandpass for MAD and SAD experiments, as well as a reduction in the vertical focal spot size to 30 microns. Great care must now be taken centring crystals. The beam flux density is increased by a factor of 3 - 4. It is now recommended to  use 15 - 25% of the beam intensity for your experiments (less for S-SAD).

The beamline is being further upgraded with the addition of a KETEK fluorescence detector (to replace the Roentec detector) and a XIA controller which will enable you to save the fluoresence spectrum.  This Equipment will be installed during January 2014.

We are currently looking into a further increase in data processing power, as this is a limiting factor with the upgraded PILATUS detector.

The MX beamline User Liaison Committee met in April 2013. The minutes of the meeting (in French) can be found . Thanks to Marie-Helene Ledu and Valerie Biou for their rapidity in writing and publishing the minutes.


Dewar Handling:

Dewars should be sent via your prefered shipper to :

[B. Guimaraes or P. Legrand or A. Thompson]

Synchrotron SOLEIL

L'Orme des Merisiers, BP 48

91192 Gif sur Yvette.

It is recommended that you notify your local contact of shipping. The Dewars will be filled on arrival. Return shipping must be pre-paid.

>> SOLEIL Users << 



Below is a list of the most recurrent problems:

SmarGon not responding



  1. After a crash, verify if the device SGONAXIS is on STANDBY or ALARM
    • STANDBY : there is no problem with the gonio; proceed with the other checking such as robot or else
    • ALARM : there is indeed a problem with the SmarGon; proceed with the step 2
  2. Visually verify that the Chi is still on its frame
    • Yes : proceed with step 3
    • No : call 9746 and don't do anything else
  3. Visually verify if a sample is mounted on the goniometer head
    • Yes : there are two possible cases depending on whereas you were experimenting in the Robot mode or in the Manual mode. In both cases, enter in the experimental hutch and remove manually the sample. Then follow the procedures for the Robot (steps 4 and 5) or for other cases (step 6 and on).
    • No : proceed with step 6


Case of the Robot mode

  1. Make a search of the experimental hutch
  2. Turn the key to be on remote control before proceeding with step 6


Next actions

  1. From within MXCuBE, run the sub-menu [Proxima 1] > [Reference Gonio]
  2. Start back experiments as if nothing happened


Information is available on the beamline for most problems. Here we resume the most common problems.

CRYSTAL CENTRING: -- When you have clicked the "Loading" button in MXCUBE, the goniometer moves to the crystal loading position. The goniometer X-Y-Z stage should, at this stage, be "untilted" (parallel to the goniometer phi axis). The stage can become tilted if the reference position has been lost (which occurs after trying to centre a pin that is too long). If it is tilted, refer to your local contact for the procedure to realign it. Do not attempt to transfer or centre your crystal until this has been done,

ROBOT: -- The robot is a relatively new piece of equipment, and as such needs particular care. A short user manual will appear on this page soon.  Specific issues are : -

  • a) When the hutch search is broken, the control of the robot from the GUI is lost. It is necessary to push the "RESET" button to re-establish contact (for example to open the lid to recover crystals at the end of the shift). THE OPENING OF THE LID CAN ONLY BE DONE WHEN THE ROBOT ARM IS IN ITS HOME POSITION, AND THE HUTCH CLOSED AND INTERLOCKED.
  • b) The robot will stop moving WITH THE SAMPLE STUCK INSIDE THE GRIPPER if the user attempts to change the beamline configuration while the robot is in operation (for example starting to centre a crystal before the robot has finished loading a sample. BE PATIENT.
  • c) It is always wise to watch what the robot is doing but the video image of the robot on the data processing computer SLOWS DOWN THE NETWORK AND HENCE DATA ANALYSIS ON THIS COMPUTER.  Close the window if you have processing speed problems.

In the case of difficulty with the MXCube interface, the "restart data collection software" is available on the Desktop of the Raid927 computer on the beamline. Instructions of how to re-start all data collection software are posted on the wall by the Raid927 terminal.

If PILATUS data collection stops in the middle of a data set, you should restart the data collection at the last but one image, being careful to adjust the data collection run number (if required), image start number, and angle start number. This happens very rarely.

Note that inverse beam data collections are sequences of queued data collections that run one after the other - if you need to "ABORT" an inverse beam data collection, you must ABORT, individually, every collection in the sequence. (For example you are collecting 500 images in "inverse" block sizes of 50. This is then a total of 2 x 500 = 1000 images collected in blocks of 50 i.e. 20 data collections that must be aborted!) We strongly advise that you close the experimental shutter on aborting any inverse beam data collection, so that there is no chance of your crystal being accidentally exposed.   

Absorption edge scanning is via the "Roentec" and "Escan" functions, accessible by icons on the Raid927 desktop. Typically 5% transmission will be used for an absorption edge scan, but do check the measured count rate in the Roentec window (should be less than 30 kHz).

In case of difficulty with other equipment, ring your local contact or the experimental hall co-ordinator (97 97). 


BAG teams are expected to overlap their beamtime on the beamline (for example for 30 minutes) in order to help the next team get started. The local contact cannot be on hand 24 hours a day, and it is not reasonable to expect a local contact to come to work outside normal hours in order to show the next group how to use the beamline. 

Users manual

Data Analysis :

The xdsme scripts, written and maintained by Pierre Legrand, will semi-automatically integrate data. Downloads and operating instructions can be found by following this link.

Perform your experiments using the MXCuBE software:

The MxCuBE software allows Users to interact with the beamline hardware components and provides automated methods to perform MX experiments on the beamline. 


Beamline projects

To be coming

Internal collaboration



This scientific section was created in 2006 with all the scientists involved in biological sciences at SOLEIL. Being strongly interdisciplinary and at the junction where scientific measurement and project meet, the Heliobio group focuses on the development of innovative methods in order to help advance the most difficult « external user » research. This is of critical importance in the highly competitive and labour intensive field of the biological sciences: by allowing the study of the widest range of samples, these methodological improvements can make the difference between success and failure, or publishing the first or second. Several very important publications have resulted from this symbiosis between beamline scientist and user at SOLEIL.



External collaboration




Laboratoire Inter-Universitaire des Systèmes Atmosphériques