The Best Liquid Scintillator For Continuous-Flow Measurement

In formulating liquid scintillators for flow application, the most important considerations must be fluidity and miscibility; counting performance is secondary. Above all else and for obvious reason, we must be able to achieve flow, not merely of the scintillator solution but of the mixture of scintillator with mobile phase. This is a rather different situation from discrete sample counting in vials where performance - efficiencies and backgrounds - come first, with fluidity and ease of mixing being of decidedly less consequence.

Another issue is that of chemiluminescence, a frequent unwanted by-product of the mixing of scintillator solution with many mobile phases. Despite circuitry intended to cope with small amounts of chemiluminescence and/or phosphorescence, the appearance of these phenomena in consequential amounts is more or less catastrophic in a flow situation but often relatively tolerable with a liquid scintillation counter where there is time for dark adaptation and cooling.

These points are cited to make the case that flow measurement is different enough from discrete sample measurement so that the best scintillator formulations for one type of work are not likely best for the other. Just because an investigator has had good success counting fractions in a liquid scintillation counter does not mean that the same scintillator will be best for flow measurement.

Once that premise is accepted, it is well to think about the scintillator:mobile phase ratio. With a liquid scintillation counter, the interrelation of one sample with the next is simplified if the sample makeup is held constant. It is not uncommon to find that day after day, month after month, year after year, an investigator will pipette 1 ml of sample solution into 10 ml of scintillator solution in a counting vial, shake the mixture together, and place the vial in the counter for a five- or ten-minute count. It's a good formulation, it works, the counting vial is always half filled, there aren't problems, why make changes?

It isn't quite the same with a flow-through detector. Other things being equal, a 5:1 scintillator:eluate ratio results in half the counting time that there would have been had the ratio been 2:1. But, a 2:1 ratio means that if the mobile phase has mild quenching properties, as it almost always does, the quencher, i.e., the eluate, makes up a larger percentage of the counting solution. The counting efficiency is probably less than it would have been at 5:1. So, we find ourselves with a dilemma: a higher ratio means less quenching which is good, but it also means less counting time, which is bad.

(In passing, we should note that in the measurements that we make, the substance actually being measured is present in such low concentrations that it rarely affects performance; the mobile phase is the quencher while the test substance contributes the counts.)

How are the problems posed above overcome? We know that different isotopes are affected differently by quenching agents; the amount of quencher that reduces 3H counting efficiency by 50% might lower 14C counting efficiency by only 10%. With this being a rather general phenomenon, it is usual to employ more scintillator for 3H counting than for 14C. If there are no other indicators as a starting point, with conventional scintillators we suggest a 3:1 scintillator:eluate ratio for 14C and a 4:1, ratio for 3H measurement. For 3H/14C double-isotope counting, the 3H ratio is the one to use. With a more modern scintillator such as IN-FLOW 2:1, one might advantageously use a 2:1 ratio for 3H and 1:1 for 14C

Why not even less? That would save scintillator cost, lessen disposal problems, and lengthen the residence time that each peak is in the detector thereby increasing the number of counts recorded. Unfortunately, this is one of those good ideas that usually does not work. Mixing is often poor, stiff gels form, the high mobile phase content lowers the counting efficiency, and closely spaced peaks are not adequately separated.

But, which scintillator, and can we fine-tune the ratio? The only way is to test. Here, the liquid scintillation counter is useful. Some of the isotope to be measured is added to a small sample (10-25 cc) of the mobile phase. If a double-label experiment is contemplated, only 3H need be added; if a gradient is involved, both ends should be tested. The quantity of isotope is almost immaterial and need not be known as long as it is sufficient to provide an activity level of several thousand cpm/ml. For each scintillator to be tested, pipette 1 ml of mobile phase into each of three counting vials. Then, to each vial add scintillator in the ratios of 1:1, 2:1, 3:1, and 4:1 for 14C or 2:1, 3:1, 4:1, 5:1 for 3H.

Gently mix the contents of each vial; remember that a flow-through detector does not have a vigorous mixing element. Count each vial for a minute or two until you feel that you've obtained sufficient counts to know if there are significant differences in the count rates. Bear in mind that these samples have different volumes with different geometries and so they must not be compared too closely.

Which is best? If in any series you see chemiluminescence, that scintillator must be forgotten. Similarly, stiff emulsions should immediately rule out a scintillator or a particular ratio. If any series clearly outperforms the others, that scintillator should be chosen. If any series underperforms the others, that scintillator is best forgotten.

Within a series, which is the best ratio? The best ratio is that represented by the vial whose count rate divided by its volume is highest. For example, if 2:1, 3:1, and 4:1 are tested with 14C, and the count rates found to be 6,000, 9,000, and 10,000 cpm respectively, then 6000/3 = 2000, 9000/4 = 2250, and 10000/5 = 2000; the 3:1 ratio is best.

Before continuing, are such numbers believable? They are. Bear in mind that a 2:1 ratio means that 33% of the mix is potentially a mild quencher. At 3:1 the mobile phase still makes up 25% of the solution while at 4:1 it drops to 20%. We oughtn't be surprised to see the count rate increase as the percent of quencher declines, but as each additional increment of scintillator has less and less dilution effect, increases in count rate taper off until ultimately the rate remains constant.

But, what scintillators should be tested? You must look to the mobile phase. If you are working with high salt content, or a highly buffered solution, there are special mixtures for such solutions. Counting solutions have been developed for mobile phases containing a high content of acetonitrile. There are biodegradable scintillator solutions when disposal might be a problem and inexpensive ones when cost containment is of interest. You are likely to find that only formulations specifically intended for continuous-flow application will prove satisfactory; the others will gel and not be pumpable somewhere along the line.

We propose counting perhaps a dozen samples, each for a minute or two. We know that some people will object; 30 minutes to learn something useful is too much. Unless you have experience with the kinds of solutions being employed, there really is no good alternative. But, with a little understanding of principles and a simple program for test, it should be possible to select a good, if not the best, protocol for flow-through measurement with only modest effort