Our atmospheric sampling and trapping efficiency is 100 times better because we got rid of differential pumping section of the instrument.  Almost all commercial instruments are differentially pumped.  In other words there is a section in between the inlet and the first quadrupole chamber that is mechanical pumped to roughly 1 or 2 Torr.

At that pressure, the analyte ions follow the flow of the carrier gas.  Generally the gas flow through the roughing pump is 1000 to 10,000 times the flow of gas into the first multipole ion guide chamber.  Therefore, 1 out of every 1000 or 10,000 ions makes into the first ion guide.  That is a colossal waste of analyte.

But they put special ion optics in that region to improve ion transmission.  

For example the S-lens on the Thermo Velos improves ion transmission by a factor of 10 according to their advertisement!  Great! Instead of losing 10,000 ions to the roughing pump you now lose only a 1000 for every ion sampled into the first guide.  Congratulations, this is the equivalent of putting a Band-Aid® on an axe wound.  In a word, “ineffectual”.  It does not solve the problem, and it is an expensive solution.

What is your solution to the problem of differential pumping?

Our solution was to remove the differential pumping completely and step the expansion of the ions into vacuum. We use a flow limiting orifice (100 mm diameter) to expand the analyte ions into a plenum chamber.  The pressure in the plenum chamber can be adjusted with a micrometer that sets the spacing between a cylinder and exit orifice.  There is only one inlet and one outlet to the plenum chamber.  Every gas molecule that goes into the plenum chamber has to exit through that exit orifice and then be injected into the quadrupole.  Unless an analyte ion impacts on one of the metal surfaces and neutralizes, it has to exit into the ion guide where, unlike the carrier gas, it is caught by the multipole fields.  The pressure in the plenum chamber is high enough (~300 mTorr) that the stopping distance of intact protein ions of any size is measured in millimeters.  Under those conditions the ions quickly entrain in the Laminar gas flow and follow it through the chamber.  To impact on the walls the ions have to diffuse through the gas and impact on the surface, a very slow process for such large ions at such high pressure.  Consequently, the transmission efficiency through the chamber should be very high. Because the pumping speed of the turbo pump on the first ion guide is 250 L/s, the pressure in the first guide chamber is 5 mTorr regardless of the spacing between the cylinder and the orifice or the pressure in the plenum chamber.

So how effective is it?
The inlet and trapping system was designed to handle ions far larger than singly charged proteins.  To demonstrate that, we created polydisperse aerosols of urea and used a differential mobility analyzer to create size selected monodisperse singly charged nanoparticles.  Faraday detection was used to measure the number density of the singly charged monodisperse nanoparticles before the inlet.  The particles were trapped in the linear quadrupole at a point just before the exit end cap electrode and then ejected on demand into a faraday plate.  The sampling and trapping efficiency was calculated from the volumetric flow rate of the inlet, amplifier gain, sampling time, and the time integrated current at the detector.  The trapping efficiency as a function of particle size is plotted in the figure below.

Millions of singly charged particulate ions with diameters that range from 10 to 200 nm were sampled from the atmosphere, trapped and then Faraday detected.  The mass range of this measurement is from ~400 kDa to 3 MDa.  The sampling and trapping efficiency range from ~4.5 % to 1 % with the efficiency increasing with decreasing particle size.  The sampling and trapping efficiency is based on the stopping distance inside the plenum chamber.  Because diffusion is negligible for such large ions, the stopping distance defines the probability of the particulate ion impacting on the wall and neutralizing.  Smaller ions have a lower probability of impact during travel through the plenum chamber and thus a greater sampling and trapping efficiency. The ~4.5% efficiency we have observed is likely to be even higher in mass range of small molecules and peptides.  Even if it remains unchanged, we have demonstrated sampling and trapping efficiency of roughly 1 part in 20 as opposed to 1 part in 1000 to 10,000.  Now that is a significant improvement.