Dry Run of Technical Evaluation

Purpose, Procedure, and Overview of Results

The purpose of the "Dry Run" of our technical evaluation of the mineral oil samples was to find out how well we could trust our testing procedures and how well we could differentiate between oils. Oils from four samples that we have been working with were pumped into 1 gallon milk jugs and labeled by a random number. The people doing the measurements did not know which oil was which. (The oils used and the sample numbers are give in the table in the next section.) Each oil was tested for attenuation length (using "Eric's sample") with the Alabama tester, for transmission shape (using "Jen's sample") using the Cincinnati tester, for index of refraction and density (using "density sample"). The results are given below. After the tests were complete, the results were compared and compared with a known standard of Witco oil. The density and index of refraction was acceptable for all oils. For Eric, either sample 2 (Duoprime 70) or sample 4 (Witco) could have been the Witco. These two samples were clearly the leaders although sample 3 (Superla 5) was not far behind. All three were clearly better than sample 4 (Drakeol 7). For Jen, samples 3 (Duoprime 70), 1 (Superla 5), and 4 (Witco) were clearly better than sample 2 (Drakeol 7). Her ordering among the top three (Duoprime 70, Superla 5, and Witco) differed from Eric's (Duoprime 70, Witco, Superla 5).

Oils Used

Oil Name	Manufacturer	Sample - Density	Sample - Jen	Sample - Eric
Superla 5	Amoco	4	1	3
Drakeol 7	Pennreco	3	2	1
Duoprime 70	Lyondell	2	3	4
Scintillator Fluid	Witco	1	4	2

Density and Index of Refraction Measurements

Index of refraction measurements were made with the same plexiglass prism that Narumon used in the initial measurements documented on the index of refraction page. Density measurements were made by weighing a 250 ml graduated flask both empty and filled. The flask was filled such that the bottom of the oil meniscus was level with the 250 ml line. Between each measurement, the prism and the flask were rinsed with ethyl alcohol and allowed to dry. Six lines of mercury were measured. The the angle of minimum deviation was set for the Hg green line and the other five lines were measured with respect to this line.

The following table summarizes the results of the density and index of refraction tests:

The plots of the index of refraction measurements vs. wavelength can be seen by clicking on the appropriate index. The values of n_D and the dispersion are calculated from the single pole resonant model described on the index of refraction page. All measurements fit very well with this model.

Test runs of three oils were made and the results of those are summarized in the following table:

Based on the above numbers, it was concluded that the density sample 1 was the Witco oil.

Jen's Attenuation Tests

Jen has written up a report describing her "Dry Run" measurements. The results in that report are summarized here. Click on the various sample numbers to see the various runs on the individual samples. All of the tube out runs for the dry run can be seen in the attached figure. Likewise the tube empty runs can be seen in the next attached figure. As can be seen in these figures, the empty tube runs have a lot more variation that the no-tube runs.

Sample 3, when originally tested, gave unphysically good transmission above 420 nm (see attached figure; transmissions above 92% are more than can be expected). It was re-tested and found to behave more normally. However, this abnormality indicated that we cannot trust this setup to give absolute attenuation lengths. The transmissions of all four oils are plotted in the attached figure.

Eric's Attenuation Tests

Measurements were made on the four dry run test oils with the Alabama tester in the configuration described on that page. After measuring oil 2, Eric discovered that his oil transfer procedure splashed oil onto the testers lens, resulting in anomalous loss of light. Oils 3 and 4 were measured after cleaning the lens after each fill. Oils 1 and 2 were re-measured after this. However, no known samples were re-measured. That will be done after the real samples are measured. The following table gives the results of the dry run oils:

The plots of the data for each of these oils can be found it in the attached figure. Two fits are drawn on these figures: 1) an exponential fit, and 2) a second order polynomial fit.. The attenuation length from the exponential fit is given at the "fit" attenuation length in the above table. In all cases, the second order polynomial fits much better than the exponential. The zero intercept divided by the slope at zero is given as the attenuation length at zero. Likewise the fit value at 90 cm divided by the slope at 90 cm is give as the 90 cm attenuation length. All of this data normalized such that the last data point is equal to 1 is shown in the attached figure. As can be seen in this figure, oils 2 (blue) and 4 (purple) clearly differentiate themselves from oils 1 (red) and 3 (light blue). This ordering differs from that of Jen's tests.