Does SIM/Scan reduce SIM sensitivity

[MSCHemist]

Title says it all. I am doing a method using SIM and running scan at the same time. If I just do SIM would that increase sensitivity?

[fendersim]

In theory I guess it should.

However, just try it and see if you can tell any difference.

[MSCHemist]

Actually I don't. They are about the same.

[Loekie]

Hi MSChemist,

you probably would have to look at the reproducability improvements.

When you omit the sim scan you can increase the scan/dwell time in the sim scans. Longer scan/dwell times do not increase the signal intensity. However it should increase the signal to noise ratio and make the integration better and area more reproducable because you see more ions.

[lmh]

umm... whether longer times increase the signal intensity or not depends on the manufacturer and the instrument. In Agilent LC-MS instruments running Chemstation, increased time definitely does increase signal.

The signal:noise should increase with increased dwell time, but it's not linear. I think it goes up with the square root or some such thing in theory (think of standard error, which is the error on an actual estimate of a value, and goes down with the square root of the number of times you measure the value). This means that doubling dwell time increases S:N ratio by only about 40% at best, which you won't notice from looking at only a couple of runs. In practice, it often doesn't go up as much as it theoretically could.

[Loekie]

So it seems that if an agilent scientists is looking at a car longer it grows bigger. That's nice. If you want a bigger car just look at your carr for a while

[Gaetan Glauser]

We've recently measured standard compounds in MRM mode at different dwell times on a QTRAP 4000 (200ms, 50ms, 20ms, 5 ms). As mentioned in the previous posts, the signal was stable and did not decrease with reduced dwell times. Noise increased slightly with decreasing dwell times, in particular at 5 ms (about 3 times higher than at 50 or 200 ms). I always adapt the dwell time to get enough but not too many data points across chromatographic peaks. This will depend on the number of simultaneous transitions that are monitored.

[James_Ball]

Eliminating the scan while keeping the same sim dwell times would give you more scans per second which would increase the number of scans across the peaks, which could improve how well the peaks are defined. This would give better integration possibly.

I just did some work using sim/scan in combination and found that changing the sampling on the scan portion from 2 to 3 gave me a couple less scans per peak in the scan mode but increased my overall sensitivity in the scan mode but it did not change my sim sensitivity at all. Increased dwell times in either mode can help sensitivity some, but not as much as you might think, and if you sacrifice number of scans per peak for increased sensitivity you may actually degrade your results instead of improving them.

[lmh]

So it seems that if an agilent scientists is looking at a car longer it grows bigger. That's nice. If you want a bigger car just look at your carr for a while

To be fair, what seems to happen in our single quad using Chemstation actually reflects what's happening quite accurately. In SIM mode you're setting the quadrupole to pass your favourite ion, and counting how many ions come out the end. If you wait twice as long, twice as many ions come out the end. But of course more noise-events will happen during the longer counting-time too.

If you want to criticise the measurement technique of manufacturers, Thermo are on far thinner ice in their older ion-trap instruments. They give an ion count that could be construed as the total number of ions scanned out of the trap. So why do their low-capacity 3d traps of the DecaXP generation give vastly higher numbers than their high-capacity linear traps of the LTQ generation? That does hint at a sales pitch in the DecaXP period: Big number = Big sensitivity (not true!).