Subject: Re: CI problems
From: Bill Cotham
Date: 28 Oct 1996 10:01:48 -0500
John Barton wrote:
}
} Kerry M. Peru wrote:
} }
} } I would like any suggestions that would aid in correcting the following
} } problem that I have been experiencing while trying to perform some CI work.
} }
} } I operate an Autospec Q and for some reason I am having difficulties with
} } achieving a clean CI spectrum. Here are a list of the problems encountered
} } and remedial actions taken to date:
} }
} } Upon introducing the CI gas (I have tried Ammonia, Methane and Isobutane) a
} } large background is apparent ie: a peak at every mass with emphasis at the
} } low end of the scan range (I am scanning from 60-500 daltons). I have tried
} } a full range of the emission current settings and eV settings but the
} } background is still abundant enough to mask the analytes of interest and
} } create very poor apparent sensitivity. My source temperature is at the
} } lowest it can maintain (130C) and the source is as tight as I can make it. I
} } am definitely achieving CI conditions, as I have introduced known compounds
} } at a concentration that I are visible and the M+1 ion is the most abundant.
} }
} } I have cleaned the CI source, changed ceramics, cleaned the source housing,
} } leaked checked all lines, baked the source at 300C, but I still get this
} } large background signal. I think it is some sort of contamination (likely
} } hydrocarbon) but I do not know where it is coming from or how to get rid of
} } it. I ruled out contaminated CI gases, as I mentioned previously, 3
} } different gases were used.
} }
} } A direct e-mail to me at PeruK@nhrisv.nhrc.sk.doe.ca of any suggestions
} } would be greatly appreciated.
} }
} } Kerry M. Peru
} } National Hydrology Research Centre
} } Environment Canada
} } 11-Innovation Blvd.,
} } Saskatoon, Sk.,
} } S7N 3H5
} } ph: 306-975-4206
} } Fax: 306-975-5143
}
} I also have an AutoSpec-Q, and do a lot (2000 samples/year) of C.I.
} using ammonia as a reagent gas, so I have some observations which may
} help. There are usually quite a few background peaks, but these are of
} low intensity and are always swamped by analyte peaks if the source is
} working efficiently in C.I. mode.
}
} Try introducing ~ 2 microlitres of amyl actetate into the septum
} reservoir, and use ammonia as the reagent gas. Set the ammonia pressure
} in the source housing to about 5e-5 mbar, and tune to optimise the peak
} at m/z 148 (M+NH4). If things are going right, this peak should be at
} least a factor of 20 higher than the peak at m/z 70. In fact, I find
} that m/z 70 is barely detectable, and the 148 peak is off scale with the
} multiplier at 220 volts.
}
} If no amount of tuning helps, then check the seal between the solids
} probe and the source block. A chip out of the probe tip ceramic will
} cause considerable gas leakage. Also check the seal between the
} reentrant and the spring-loaded sliding tube which mates with it. I once
} found that someone had fitted a weaker spring to this tube, which was
} enough to degrade the CI performance through leakage.
}
} You do not say what type of solids probe you have. The water-cooled
} probe has a two-piece ceramic tip. Part of this tip is made of
} machinable ceramic (macor), and part out of boron nitride. These
} materials can absorb certain chemicals, which then bleed out over a long
} period of time. Try baking the probe at its highest temperature, or
} replace the ceramic tip. If you replace it, it will still need baking,
} as the macor absorbs CO2 and the boron nitride absorbs water in the
} atmosphere.
}
} One possible source of contamination is the tubing used to connect the
} gas bottles to the instrument. Make sure all the tubing is GC-grade
} stainless steel, and all valves are also stainless steel, and preferably
} packless.
}
} I hope you find these comments useful
}
} Good luck!
}
} John Barton
} Chemistry Dept,
} Imperial College,
} LONDON, UK
Just a quick comment about the ceraminc probe tips. John is right about
the two piece tips causing some problems. We have switched to a one
piece tip made by SIS which works much better. Usually no baking or
degassing required at all. I also concur on his comments about CI.
With no analyte, you will almost always see the low mass "grass" in a CI
spectrum.
--
Bill Cotham, Ph.D.
Mass Spectrometry Laboratory
Dept. of Chemistry and Biochemistry
University of South Carolina
Columbia, SC 29208
803-777-2039 (voice)
803-777-9521 (fax)
Subject: Re: question on TOF data acquisition electronics
From: Martin@nezumi.demon.co.uk (Martin Tom Brown)
Date: 28 Oct 1996 23:15:50 -0500
In article <54tq0f$dk0@acmez.gatech.edu> wyu@genetics.com "Wen Yu" writes:
} For MALDI-TOF instrument running peptide and protein, you'd need a
} trasient recorder that, at minimum, can:
}
} 1) sample the analo signal every 5-10 ns, or 100-200 MHz
} 2) record 48K data points from a single transient
} 3) digitize the signal at 8 bit (256 steps) vertical (amplitude) resolution.
Seems reasonable for a basic system spec.
} An ideal data acquisition system should have:
}
} 1) sampling rate of 500ps/sample (2 GHz) or higher
} 2) record 48K or more data points
} 3) 12 bit vertical (amplitude) resolution
This is probably both over and under specified.
The higher the sampling rate you use the more data points are needed.
Also at these very high data rates it doesn't make sense to digitise
at a higher 12 bit resolution, you can probably get by with 6 bits.
It's the technique of oversampling with lower resolution ADC's.
} 4) good analog circuit with minimum ringing or impedence mismatch
This is down to the electronics between your detector and the card,
even the correct choice of cables and plug sockets make a difference.
} 5) high data transfer rate from digitizer to host computer (better than 3 Hz)
I expect you want it 10^6 times faster...
} As far as I know, no commercial vendor can supply so called
} "ideal" data acquisition system. Let me kown if you find otherwise.
There are a couple of even PC based cards which come quite close to
the optimum which can be acheived with present (affordable) hardware.
Regards,
--
Martin Brown __ CIS: 71651,470
Scientific Software Consultancy /^,,)__/