In a series of posts, we are going to talk about Mass Spectrometry.

1. Introduction-The dif­fer­ent con­fig­u­ra­tions and the Elec­tron Impact process
2. What types of mass ana­lyz­ers are there?
3. What type of detec­tors are there?
4. What types of analy­sis can be done?
5. How do you read the output?
6. How do they come to a qual­i­ta­tive mea­sure using software?
7. How do they quan­ti­tate the results?
8. Do you need chro­matog­ra­phy if you are using Mass Spectrometry?
9. Other top­ics of inter­est about GC-MS

One of the lim­i­ta­tions of any sort of detec­tion sys­tem is the con­ver­sion of that response to a com­put­er­ized sys­tem. This is called the Ana­log to Dig­i­tal Con­ver­sion (ADC) prob­lem. A response works in an ana­log domain, but com­put­ers work only in the dig­i­tal domain. So, there must be a con­ver­sion. The ques­tion becomes sam­pling rate or the num­ber of dis­crete data points that are taken. The more data points we col­lect the closer the dig­i­tal sig­nal is to the ana­log and the bet­ter the con­ver­sion is to the true sig­nal. This whole con­cept is the called the data acqui­si­tion rate. This is a vari­able that can be con­trolled by soft­ware or hard­ware. One way we can mea­sure is this rate is typ­i­cally referred to as the num­ber of points across (really along) the peak.

We need at least 15 to 20 points across a chro­mato­graphic peak for good quan­ti­ta­tion. If  we have fewer points we will not be able to describe the peak ade­quately and may  lose infor­ma­tion. For exam­ple, we may miss the top of the peak. We may miss a dou­blet or some other sig­nal of co-elution. We can increase the num­ber of points across the peak by scan­ning faster how­ever data qual­ity is sac­ri­ficed. If we increase the num­ber of points across the peak we must  increase our scan speed, so the time that we dwell or stay on and look at our mass of inter­est is shorter. A shorter dwell time neg­a­tively affects sig­nal to noise (s/n). The­o­ret­i­cally the s/n increases with the square root of the increased dwell time.

Although the above shows the con­cept on LC-MS-MS, the prin­ci­ple holds true for GC-MS as well.

In any form of MS work, we have to remem­ber we are “weigh­ing” mass, not its weight. These two con­cepts are not the same. Some con­cepts are help­ful as we are about to look at data. We want to define some terms so that we know how the var­i­ous axes.

• The uni­ver­sal unit for mass is the Dal­ton (Da). A Dal­ton is equal to 1/12 the mass of the iso­tope of car­bon 12. Stated dif­fer­ently 1 Da=1 atomic mass unit (amu).
• The mag­ni­tude of the charge is an elec­tron. It is denoted as “z.” “z” is the charge of the elec­tron generally.
• The m/z rep­re­sents the dal­tons per unit of charge. This is the set stan­dard that every­thing else is related to through­out our analysis.

The ulti­mate result of a hyphen­ated tech­niques such as Mass Spec­trom­e­try when it is cou­pled with some chro­mato­graphic tech­nique such as Gas Chro­matog­ra­phy or Liq­uid Chro­matog­ra­phy  is gen­er­ally called a mass spec­trum (click on to get the IUAPC). It is not a graph.

The first result will typ­i­cally be a Total Ion (Cur­rent) Chro­matogram. Again, it is not a graph. It is a plot along an x-axis and y-axis of the total ion cur­rent vs. reten­tion time respec­tively obtained from a chro­matog­ra­phy exper­i­ment cou­pled with mass detec­tion. It is a snap shot of typ­i­cally a very large win­dow often of sev­eral hun­dred mass-to-charge units. (Source: Hand­book of GC/MS: fun­da­men­tals and appli­ca­tions By Hans-Joachim Hübschmann)

TIC data ver­sus SIM data-notice with SIM we lose some detail of the over­all picture

Here is a TIC from a DUID-Marijuana case.

A TIC from a real DUID prosecution

Diag­nos­tic SIMS in a real DUID-Marijuana case

Again, real SIM analy­sis in a DUID-MJ case

A close up of two SIMS on a real DUID-MJ case

Once we have the TIC or SIM we can con­vert it to a mass spec­trum that will then be com­pared against known stan­dards (more on this later). A mass spec­trum is like the below:

A mass spectrum

The x-axis is our old m/z. The y-axis of a mass spec­trum rep­re­sents sig­nal inten­sity of the ions. In most forms of mass spec­trom­e­try, the inten­sity of ion cur­rent mea­sured by the spec­trom­e­ter does not accu­rately rep­re­sent rel­a­tive abun­dance, but cor­re­lates loosely with it. There­fore it is com­mon to label the y-axis with “arbi­trary units.” The high­est inten­sity ion is assigned 100 per­cent with all oth­ers mea­sured rel­a­tive to it.

The very first part we should be look­ing for is to try to iden­tify the mol­e­c­u­lar ion. A mol­e­c­u­lar ion is an ion formed by the removal from (pos­i­tive ions) or addi­tion to (neg­a­tive ions) a mol­e­cule of one or more elec­trons with­out frag­men­ta­tion of the mol­e­c­u­lar struc­ture. The mass of this ion cor­re­sponds to the sum of the masses of the most abun­dant nat­u­rally occur­ring iso­topes of the var­i­ous atoms that make up the mol­e­cule (with a cor­rec­tion for the masses of the electron(s) lost or gained). If present then it cor­re­lates to the mol­e­c­u­lar mass of the compound.

Think of it as a bound­ary. In EI-based GC-MS work, if the mol­e­c­u­lar ion is present and if we have done good qual­ity work, then we should never have a spec­trum whose high­est value on the x-axis exceeds the reported mol­e­c­u­lar mass.

For exam­ple, let’s use the case of ethanol. What I do is when I have data and a call by the ana­lyst, I first take the con­clu­sion sheet where the ana­lyst says it is ethanol (to the con­clu­sion of every­thing else in the uni­verse), then I go to Wikipedia and some other source to find out the mol­e­c­u­lar mass. In the case of ethanol, it is reported as 46.07 g/mol or there abouts. I next go straight to the TIC, and then the scan or the SIM and deter­mine whether or not I have iso­topes reported way out­side of that 46.07 range. So, for exam­ple, if I have a spec­trum that pro­duces ions at 100, then I know we need to do much fur­ther review and I am well on the track to prove the analyst’s call as wrong. Another very sim­ple exam­ple is in the case of a spec­trum that indi­cates a mol­e­c­u­lar weight of 18. This mol­e­cule must only con­tain frag­ments no greater than that of oxy­gen which has a weight of 16, hydro­gen, which has a weight of 1, car­bon which has a weight of 12, and/or nitro­gen which has a weight of 14. We obvi­ously turn to H2O which is 2(1)+16=18. This should be con­firmed by ions at masses 17 and 16 which rep­re­sent the frag­men­ta­tion and log­i­cal neu­tral loss of HO and O.

There is a caveat to this. As GCs will only accept volatile organic com­pounds (VOCs), if the sam­ple and in par­tic­u­lar the ana­lyte of inter­est is not a VOC, then we need to chem­i­cal change the sam­ple (called deriva­ti­za­tion) to make it capa­ble of being ana­lyzed by GC. What we have to remem­ber is that not all analy­sis on an EI-based MS will result in a mol­e­c­u­lar ion.

In sum­mary, when we do our analy­sis and we do get the mol­e­c­u­lar ion it is an impor­tant part of elu­ci­dat­ing the unknown as it estab­lishes the mol­e­c­u­lar weight and then often the ele­men­tal com­po­si­tion of the mol­e­cule (i.e., the nitro­gen rule). The EI frag­men­ta­tion and the cor­re­spond­ing EI frag­ment ions indi­cate the pieces of which the mol­e­c­u­lar ion is com­posed. This frag­men­ta­tion is through a process called log­i­cal neu­tral loss analy­sis. It is sim­ply a way of putting all of the pieces back to together again.

EI-based GC-MS is not unlike the kid’s fariy tale of Humpty Dumpty, but instead of the King’s Horses and the King’s Men, it is the mass spec­tro­scopist. Can she or he put the ions all back together again?