I have recently been given a project to bring up the gc on site for waste solvent analysis. I have installed the correct components(hp-1 column) and have ran an unscientific test to see if it is working properly. I need to know how to proceed from here on out(calibration, method development, etc.)
By Bruce Freeman on Wednesday, October 27, 1999 - 06:42 am:
Are you working from an existing method of analysis? If so, does it specify things like injection port liner, injection mode, oven temperature, gas flows, system suitability tests, etc.? If so, you may simply be able to apply the method and get the results you need.
If not then you are going to have to provide the expertise yourself to get the method running properly. While you can ask questions at a forum like this, it is impossible for anyone to provide useful answers without a comprehensive understanding of the sample and the assay requirements.
As a first step you may be better off to see whether you can get that information locally -- ask your supervisor or a coworker. If you can't get the help you need that way, then be prepared to thoroughly describe the assay you need to achieve, especially the analytes and the sample matrix.
By Anonymous on Wednesday, October 27, 1999 - 02:18 pm:
From that I remember, the HP-1 column is similar to a DB-1. You may have a problem with this column if you are trying to separate low boiling solvents. I would recommend using something like a Restek RTx-624 (6% cyanopropyl / 94% dimethyl polysiloxane), 0.53mm x 60m, 3.0um, or a Supelcowax (bonded Carbowax) 0.53mm x 60 M, 2.0 um.
The first thing that you need to do is to identify all of the solvents that will be present and develop a separation. In using 0.53 mm capillary columns, you may be able to get away with injecting neat mixtures / samples with using a large split ratio and lower detector sensitivity. You will have to pay close attention to your peaks shapes and heights to see that you are not overloading the column or the detector and data system. If your solvents are all high boiling and you can resolve them from a low boiling solvent such as acetone, methanol or diethyl ether, you can dissolve your standards / samples in these solvents. Concentration? You will have to experiment.
Make up a mixture of all of your solvents, and determine the conditions separate them. Remember that some solvents will be mixtures of a number of compounds. Verify which peaks are which by injecting single solvents. (Hope you have an autosampler.) As far as instrumental conditions, a good starting point could be: Injector 200 C, Detector between 220 and 240 C, Oven 35 C (60 C for Supelcowax) for 4 min., 35 to 200 or 220 C at 10 C / min., hold at final temp...4 min.?, 10 min.?. You will have to see if anything is still eluting off the column. A rule of thumb, you always want to keep the detector at a temperature that's about 20 C higher that any other part of your system. Try optimizing by changing the Oven Rate. You may want to change the final temp. A column vendor from which you purchased your column should be able to provide you with good guidance.
Once you establish a separation, you would want to construct a calibration curve for each component of interest over the concentration range you are will be looking at. In the process of doing this you should establish the method detection limit (MDL) and the limit of quantitation (LOQ). Different groups may define these different ways. I use a signal of three times the baseline noise as the MDL and ten times the noise as the LOQ.
Then you would want to determine the instrument precision of the method by injecting a minimum of six standards and calculating the average and %RSD.
This is a good starting point and a minimum of what I would do. There are more things to consider in method validation (which include the things mentioned above). You will have to decide the audience you are addressing and the amount of validation you need to do. For example are you addressing / reporting to a regulatory agency, or are these results for internal guidance only. This is will determine what you need to do.
For more details on method validation look into FDA, USP, OHSA, NIOSH, or EPA documents and see how you can apply their guidelines to your method / requirements. You will find that these documents contain guidelines universal to any good method irrespective of who you are reporting to.
By Anonymous on Thursday, October 28, 1999 - 10:59 am:
From my experience, we use the same programs on our DB-1's as on our HP-1's, EXCEPT that the HP-1 temperatures are lower by 20°C.
By Anonymous on Thursday, October 28, 1999 - 01:16 pm:
HP-1 and DB-1 (also SPB-1 and Rtx-1, etc.) are identical stationary phases and, given the same id, df and length, should respond identically.
By Anonymous on Friday, June 25, 2004 - 01:08 am:
I m new with the GC. Anyone can tell me how to set the Split ratio = 1/50 when flow rate is 1mL/min for capillary column manually.
Posting is currently disabled in this topic. Contact your discussion moderator for more information.