You need a high-pressure, ultra low volume (~ 1ul) 2-position valve with external contact closure control. For HPLC / LC-MS applications, the valve should have 1/16" ports and be connected between the column outlet and the detector inlet using a SS line. A properly setup LC-MS system would most likely incorporate a UV/VIS (preferable a DAD) detector in-series with the MS detector. The diverter valve would be positioned after the UV/VIS detector and before the MS inlet. This would allow you to divert any inappropriate samples AWAY from the MS source (to waste), all in-line, without needing to disconnect anything. *Note: I prefer to install a second "low-tech" isocratic pump in-line with the Diverter Valve. This enables the 2nd pump to maintain flow of mobile phase to the MS source during the times when the diverter valve takes the flow off-line (reducing any disturbance of the MS signal and also keeping the needle cool so it does not spark or heat up when the liquid flow is stopped).

The two possible positions of the valve [NORMAL, In-Line / DIVERTED, Off-Line to Waste] can be easily automated using any time-programmable contact closure within the method. If you use your Agilent 1100 HPLC to control the valve, the you will need to also purchase and install one G1351A ECI board too for remote control. You simply activate the contact (through your Analyst Software) to toggle the valve, then activate it a second time to resume normal flow.

My company, Chiralizer Services specializes in automated HPLC / LC-MS valves and has been providing these types of products since 1993 for these exact applications.

Two of our products that may interest you:

• The LC-MS Watchdog (tm), Model CS3040: A fully automated, "hands-off" system used mostly with open-access systems [ http://www.chiralizer.com/uploads/8/9/7/7/89770565/lc-ms_contamination_minimizer_brochure_page.pdf ]
• The Column Swapper (tm), Model CS3010 is an "Alternating Column Regeneration System" which can be ordered in a slightly different configuration for use as a basic LC-MS diverter valve: [ http://www.chiralizer.com/uploads/8/9/7/7/89770565/column_swapper_brochure_page.pdf ]

Thank you Bill

What Bill said is the high-tech way and appropriate for serious researchers especially in regulated environment. For university labs there is a low tech way---a simple low dispersion volume T-connector right on top of the sprayer needle. One arm to the needle, one arm to the waste. You change the diameter and length of the waste arm to adjust the diversion ratio (if your waste jar is far away you can couple the end of the waste line to a large-diameter tubing that doesn't generate back pressure). You can have a rough estimate of the ratio by collecting and measuring the two outlet flows, but bear in mind that the ratio do change during the gradient run and under different flow rates, so it's not "reportable".

This is the valve I installed on my 4000. You may seach www.vici.com for details. Rheodyn has a good valve as well. If you are short in budget, you can try ebay. Another way I called a "poor man switching valve". Before or after your peak, you set a period where your DP is 0 and ISV is 0 volt. With no voltage, any compounds would not be charged and not going to you quadrupole and going to waste along with your mobile phase spray. Of course, it will not prevent matrix deposit on the curtain plate.

also try this youtube file https://www.youtube.com/watch?v=IZUCgfvgPyU from Sciex