[RASMB] AUC interference concentration limits

Wed Oct 21 08:29:15 PDT 2015

Hi Kristian

These are all good points which you make. It is pleasing to know that 'spacers' which will correctly locate the 2/3rds plane of short path-length centerpeices are available. I'm not sure that everyone appreciates that simply loading a 3mm (or whatever) centerpiece in the mid-plane of the cell* will place the 2/3rds plane of that centerpiece so far out from its geometrically correct position that the Svensson equation cannot even be applied to see what the error is.

One other general point. I'm sure most folk do appreciate this, but it is not just being pedantic to note that at high solute gradients it is not just that the fringes are lost, but the light rays which generate that information are being refracted outside the angular aperture of the optical system used. In the days, years back, when I had a MSE Mk II AUC which had an open 'out in the air' optical system I used to demonstrate to students the effects of closing down that aperture by cardboard cut-outs. Of course the effect is not 'all or none' - the geometry of the effective aperture is not simply defined, and I can easily see that some residual information, leading to faint or very faint fringes might be found, as you describe.

Arthur

*and placing it into an instrument where the focus has been correctly located at the 2/3rds plane

On 21 Oct 2015, at 09:41, Kristian Schilling wrote:

> Hi John,
> 
> I agree that the reason for your missing data in the middle of the boundary is due to unresolved fringes in that region. This has been mentioned before, but here are some more considerations:
> 
> Your image appears much too bright for me. Some people think that an image as bright as possible and with sharp edge contrast between light and dark stripes is the optimum, but that is not true. The vertical course of light and dark follows a squared cosine function, so there are lots of grey transitions that are needed for the fourier transform. When the FT is performed on every pixel column, it can work the better, the more greyscale intermediates can stabilize the transform. When the image is too bright, the course of brightness towards the maximum within a white stripe cannot be used as the brightness is at its upper limit.
> 
> Brightness and contrast should be adjusted such that you can see as much grey as possible with only the center of the white stripes being pure white. Be aware that our eye will not judge the screen output in an objective way; it may be worthwhile to capture the image and analyse it with a graphics program to get the actual pixel values. If a pixel column apears like a wave with its maximum close to saturation, you are doing fine.
> 
> Another thing you might want to check is your camera's azimuth orientation. It looks fine when I look at your air-air region where the fringes appear perfectly horizontal. However, we have found that for high concentrations, the camera azimuth needs to be adjusted better than 0.1°. This is also related to Fourier transform, and there are some pictures, calculations and a more thorough discussion in our recent paper (Schilling, Krause: "Analysis of Antibody Aggregate Content at Extremely High Concentrations Using Sedimentation Velocity with a Novel Interference Optics"; DOI:10.1371/journal.pone.0120820). Incidentally, there are also some magnifications in that paper, showing the interior of a boundary with steep fringes. They show that there may well be detectable fringes even if they are not perceptible by eye in the screen display.
> 
> So you might be able to increase the performance of your instrument by optimizing the camera's azimuth. As a matter of fact, we have found that the best way to do this is to run scans on a highly concentrated solution at high rpm, and to turn the camera slightly between scans until the non transformed region of the boundary is as narrow as possible.
> 
> With these two measures, you might be able to achieve the theoretical limit you are referring to.
> 
> As to Wiener skewing and the 2/3 plane:
> It is crucial to focus on the 2/3 plane in order to make the third term of the Svensson equation vanish, so you can avoid the bending of the light path Borries mentioned. And I see no reason why that could not be accomplished. The Beckman technician will do this on every service of the interference system. It is important to do the focussing of the camera lens when the chamber is evacuated because the rotor will be slightly higher in that condition - which is the one we have during a measurement. And it is true that shorter centerpieces will need to be elevated within the housing so that their 2/3 plane is exactly at the same height as for other centerpieces. For this reason, we have manufactured housings for 3 and 1.5 mm centerpieces that guarantee this geometry. Please also have a look at Arthur's post on 3 mm centerpieces of Sep 19, 2006, where he demonstrates the influence of the third Svensson term in detail.
> 
> With 1.5 mm centerpieces, we have measured antibody solutions as high as 150 g/l and achieved well resolved fringes (not with the Beckman setup, but with an improved CCD and slightly higher magnification, also described in the paper above). You will not get as far as that with the standard camera, but with the adjustments described above you should at least approach the theoretical limit you cited. BTW, 1.5 mm centerpieces are the thinnest we did because it becomes impossible to shoot a 1 mm filling hole into any thinner centerpiece.
> 
> However, there is another limit: though we get well resolved fringes at such high concentrations, they will not be as steep as they should. They do not exceed a steepness greater than 45 fringes/mm. This limitation is due to the optical resolution and can only be overcome by higher magnification.
> 
> Best wishes,
> Kristian
> 
> Am 20.10.2015 um 21:32 schrieb John Sumida:
>> 
>> Dear RASMB,
>> 
>> I have been reviewing the thread posted on RASMB between September 1 2008 and September 15 2008 under the subject heading “upper concentration limit AUC”.
>> 
>> We attempted to perform an SV measurement of a sample using interference detection at a sample concentration of 17 mgs/ml.
>> 
>> At this concentration however I observe a gap in the boundary data that resulted from the spin (see bmp image attached); this behavior was not observed at lower concentrations up to 2-5 mgs/ml.
>> 
>> The gap in the data appears to be correlated with the absence of a fringe pattern in the ccd image.  Adjustment of the duration setting does not resolve the issue with the data, but it does improves the ccd image so that I can make out the fringes in the region where I’m observing the gap in the data.  Counting the fringes/mm in this region, I estimate the fringe gradient to be approximately 50 fringes/mm. This is less than the 76 fringes/mm noted in the RASMB thread but my count may be off.
>> 
>> On a separate note, since I suspect that the artifact I’m seeing is associated with the observed steep fringe gradient I had wondered if our instrument was focused at the 2/3rds plane or not.  I have been informed, by the vendor, that the process of focusing the interference optics at the 2/3rds optical plane is not something that is possible in the “real world” (their words, not mine), and that this is that this is only something that is only “theoretically” possible.  This appears to be in direct contradiction to the descriptions provided on pages 311 and 312 of Yphantis et. al. Biochemistry 1964 vol.3  (see attached) where concentrations of 12.5 and higher were measured (if my reading is correct) where measurements were performed by using the 2/3rds plane focusing.
>> 
>> Questions.
>> 
>> Is the displacement in the fringe data along the radial axis consistent with a concentration gradient that is too steep?
>> 
>> What is the best way to measure the fringe/mm value?
>> 
>> Are concentrations up to 30mgs/ml possible in a 1.2 cm cell using interference optics, and if so how should the optics be focused?
>> 
>> Thank you in advance for your time, comments and your advice.
>> 
> 
> -- 
> Nanolytics
> Gesellschaft fuer Kolloidanalytik mbH
> Dr. Kristian Schilling
> 
> Am Muehlenberg 11
> D-14476 Potsdam
> Tel: +49 331 5818360
> Fax: +49 331 5818361
> e-mail:	schilling at nanolytics.de
> Internet:  www.nanolytics.de
> _______________________________________________
> Diese E-Mail kann vertrauliche und/oder rechtlich geschützte Informationen enthalten. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrtümlich erhalten haben, informieren Sie bitte sofort den Absender und vernichten Sie diese Mail. Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser Mail ist nicht gestattet.
> 
> This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
> 
> _______________________________________________
> RASMB mailing list
> RASMB at list.rasmb.org
> http://list.rasmb.org/listinfo.cgi/rasmb-rasmb.org

Professor Arthur J Rowe
NCMH/Food Sciences
University of Nottingham
Sutton Bonington
Leics LE12 5RD UK

Tel: +44 115 9516156
arthur.rowe at nottingham.ac.uk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20151021/6bcdfda8/attachment.htm>