[RASMB] interference & meniscii

[Kristian Schilling]

Dear Holger,

we work with interference a lot and love it - feel free to call me on 
detail questions. Concerning your issues:

a)
You got me wrong on the meniscus/bottom topic. What I said was on 
absorbance. There, the only possibility to tell what bottom radius is 
sample and which is reference, you need the intensity scan. And for 
interference data, if you have slightly different loadings (like 210/200 
ul), the only way to tell which of the menicii visible in the fringe 
display is sample and which is reference is to make an absorbance scan, 
most likely in intensity mode. (Apart from the fact, of course, that you 
can tell from your filling volumes, but in case you can't be sure...)

To determine meniscus and bottom, we read them from the fringe display 
before we start scanning. I advise not to use absorbance readings for 
interference data because radial positions on both optical systems might 
not match perfectly. As a matter of fact, the camera lens in interference 
optics has to be very well adjusted for this. You might check the manual 
readings from the fringe display and compare them with an intensity scan to 
get an idea of how well your camera lens is adjusted.

We also perform an intensity scan at the end of an equilibrium run in order 
to distinguish both bottoms and to make sure which one is sample. Then, we 
take the corresponding manual reading from the fringe display. There, you 
see the smaller bottom value in the fringe-gray-transition, the higher 
value in the gray-black-transition.

Be sure to do radial calibration using the outer blade limits. As you can 
verify, the distance between the inner blades is 12 mm, between the outer 
blades 13 mm. In the region of both apertures, both laser beams pass the 
windows and cause interference patterns. In the region between fringes and 
darkness, only one laser beam can pass and no interference patterns are 
created, the intensity being that of only one beam (gray). So it is at the 
fringe/gray border where the outer blades are located and this is where we 
get the 13 mm range between 5.85 and 7.15 cm.

Also, you should do radial cal at 3k where the rotor is not stretched and 
use the readings at a higher velocity without recalibrating. The rotor 
stretching is real and should not be eliminated by recalibration.

As to clipping, if you want to use Bo Demelers Ultrascan, be sure to 
include some of the air-air-region in your scans. Other programs, to my 
knowledge, do not need these data. We prefer to clip the fringes some 0.01 
cm right to the meniscus.

b)
Our machine is from the beginning of 2000 and the timing is OK. The picture 
might become a bit misty above 50K but the fourier transformation works 
out. If you encounter problems, you might try to increase the laser 
duration from the recommended 1 degrees to, say, 1.5.

c)
The fringe displacement is correlated to concentration cia laser 
wavelength, dn/dc and optical pathlength. For sedimentation, it is the 
height of the sedimentation front, in equilibrium the "expected loading 
concentration" J0 that an be accessed with an extrapolation procedure 
suggested by Steve Harding. This procedure is part of the MStar package by 
Helmut Coelfen, the MStar on Origin platform I wrote also includes a 
refined procedure. Helmut presented the macro on the recent AUC meeting - 
Walter, have you checked yet whether to put the package on RASMB?

We use the fringe displacement/concentration relationship whenever possible 
to confirm our mass balance, checking whether we have found all the 
material we put in. This is a way to oppose the problem that you do not 
have something like a baseline absorbance in interference.

With Steve's extrapolation, we mostly get very good values for Ja even if 
we are far from meniscus depletion.

d)
There might be several effects prompting you to perform corrections on the 
sedimentation profile. With compressible solvents, for example, fringes 
will tilt significantly due to pressure differences in the sample and 
reference columns. An artifact that has to be removed and is not 
distinguishable in the final equilibrium scans. As to low weight 
contaminants, you can simulate equilibrium profiles of both components and 
judge the effect. I think that even the Beckman software offers this 
option. Unbelievable.

e)
The fringe displacement is linear with concentration without limitations. 
nb however that different materials with different dn/dc cause different 
fringe shifts and cannot be compared directly from fringe displacement. 
There is no dependence on radial position.

f)
Sure there is, but for now, let me skip this general question.

Hope the other suggestions help - enjoy interference, it's great!

Cheers,
Kristian

At 18:08 25.07.2002 +0200, you wrote:
>Dear all,
>
>recently, I started to work more with the interference optics of the Xl-I,
>and there are some points not clear to me yet:
>
>a) how do you "precisely" determine the mensiscus and bottom position of
>the cell? I started with the recommended procedure by Beckmann, but now I
>include a larger region in the scanning, but still, things are not clear.
>Kristian Schilling suggested in this forum that you can only get the
>precise meniscii positions from the intensity data, and not from the
>interference alone? (I'd rather measure that parameter than have it
>estimated.)
>
>b) There seems to have been a problem with the timing of the laser pulses
>at speeds > 50 K a couple of years ago. How's about that (our machine is
>from '99)?
>
>c) When using Lamm-equation modelling to evaluate the data, you need an
>estimate of c0; what is the best way to get (again, measure) this
>parameter? At low loading concentrations and 3-mm CPs I guess you run
>into trouble when simultaneaously estimating for both meniscii and c0?
>There is a paper by Babul-J et al. (1969), Anal Biochem 28: 216, but the
>procedure descibed in there seems to be unsuitable for low (<0.15 mg/mL of
>protein/peptide) concentrations. (And again, I'd rather fix that
>parameter, since it should not be so much of a problem to include that as
>prior knowlegde into the fitting?) Calculating the TI- and RI-noise and
>substracting it from a scan at 3 K?
>The other way of asking the question is of course how do get the baseline
>offset, when you let the SedVel continue until SedEq is established; but
>if I remember there is no way of measuring that parameter unless you can't
>deplete the meniscus (which I can't, in this special case).
>
>d) Is there a suggested procedure to test wether the change in
>concentration measured is really only due to transport of solute and not
>of anything else (or otherwise, how do you establish that chemical
>equilibrium of the low-molecular weight components really has been
>reached in the dialyzate, when you can't deplete the meniscus of your
>solute?)? I can't use absorption optics simultaneaously with the
>interference (small peptide, no absorption >240 nm).
>
>e) Is there a numerical value of fringe shifts, where the signal is not
>linear any more with the change in concentration? And of fringe
>shifts/radial position?
>
>f) Finally, is there somewhere a good standard protocol out how to plan,
>set up and evaluate experiments with interference optics?
>
>Na. Marvelling at the commplexity of things, all the best. Holger
>
>- - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>
>Holger Strauss
>
>Forschungsinstitut fuer Molekulare Pharmakologie (FMP)
>Robert-Roessle Strasse 10
>
>13125 Berlin/Germany
>
>Tel: +49 (0)30 94793 - 223 (office)
>                      - 316 (lab)
>
>Fax: +49 (0)30 94793 - 169
>
>
>- - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>
>Science is spectrum analysis; art is photosynthesis.
>
>                                                     Karl Kraus
>
>
>_______________________________________________
>RASMB mailing list
>RASMB at rasmb-email.bbri.org
>http://rasmb-email.bbri.org/mailman/listinfo/rasmb



=================================
Nanolytics
Gesellschaft fuer Kolloidanalytik mbH
Dr.  Kristian Schilling

Hauptstr. 20
D-14624 Dallgow

Tel.:		+49/3322/24200-5
Fax:		+49/3322/24200-6
e-mail:	schilling at nanolytics.de
Internet:	www.nanolytics.de