Isotope Analysis by ICP-MS
Inductively Coupled Plasma Mass Spectrometry (ICP-MS) is an analytical technique used to measure the concentration of inorganic metal species in an aqueous solution. As a general rule, ICP-MS is capable of analyzing anything to the left of the metalloid boundary on the periodic table with some exceptions and varying degrees of difficulty for each specific metal of interest. Organic compounds, complexes, and molecules are not measured and are often intentionally digested into an inorganic state in order to prevent their presence from interfering with the measurement. Even if samples are not digested, the plasma will attempt to homogenize and ionize any sample sent into the instrument, which ideally leads to the destruction of such compounds if present. The metals that are present within these compounds can be reliably measured as long as the metal of interest is not volatile in strong mineral acids. Otherwise, sampling methods specific to the metal of interest will have to be developed which can take time and be expensive.
Samples must be in acidified aqueous solution; preferably around 2% trace metal grade nitric acid, but most dilute acids will work. If your samples do not exist in an acidified aqueous state, they will have to be prepared into such a matrix. Our lab has the resources to do this for a fee. It is best to include a process blank along with your samples in order to determine the background metal levels within your experiment.
ICP-MS measurements can be very accurate and precise, but can also be problematic and expensive. The Keck Lab will strive to give you the best data possible with the samples that are given to us, but we cannot be held responsible for potential problems that may occur on the customer end such as, but not limited to:
- Instabilities due to samples that were not properly acidified or stored inappropriately by customer
- Poor results due to a lack of proper method development by customer
- Poor results due to customer end experimental failure
- Lack of desired results due to lack of customer instruction
- Dissatisfaction over time required to perform analysis
Retesting of samples can be performed for additional measurement fees. Electronic or in-person consultation can be requested before and/or after an analysis. Such consultation will be free of charge unless specifically notified otherwise beforehand and will be at the complete discretion of the Keck Lab staff.
Isotope Analysis by IRMS
A gas isotope ratio mass spectrometer (gIRMS) is a dedicated instrument for the measurement of isotope ratios of mainly C, N, O, H, and S. The ratio of the minor isotope to the major isotope (for example 13C to 12C) is measured from 4 to 6 significant figures. This kind of precision is very difficult to impossible to obtain with the more common "organic" mass spectrometers such as ion traps, FT-ICR, and even quadrupole mass spectrometers. The precision (standard deviation) of an isotope ratio measurement is theoretically limited to the square root of the number of ions counted in making the measurement. Consequently, we need to see as many ions of our analysis gas as we can in order to make the high precision measurements we are after.
I. Instrument design
The design of the gIRMS is intended to maximize the number of ions counted. Here are some aspects of an isotope ratio mass spectrometer that contrast with an organic mass spectrometer:
A gIRMS uses what is called a "tight" ion source. The chamber in which ions are formed is very small and has a very small exit slit as the sole escape for the analysis gas. The increases the residence time of the analysis gas in the vicinity of the electron beam used to generate the ions. In contrast to this, many organic mass spectrometers have a more open ion source which greatly limits (intentionally) the ionization efficiency.
By far the standard ionization technique for a gIRMS is electron impact ionization. While this technique is also used in organic mass spectrometry, organic mass spectrometers have a much wider array of ionization techniques available to them, such as electrospray, laser desorption, chemical ionization, fast atom bombardment, and many more.
A typical gIRMS will have an ionization efficiency of about 1 ion in 1000 to 2000 molecules for CO2. This means that for every 1500 molecules of CO2 introduced into the ion source one ion will be detected. Potentially more than 1 ion is generated, but there is still a long way to the detector from the ion source. This is in contrast to an organic mass spectrometer which might have an ionization efficiency on the order of 1 ion in 105 or 106 molecules. Note that ionization efficiencies are a function of a wide variety of parameters, the most fundamental of which is the ionization energy of the molecule in question. The reason why organic mass spectrometers are not designed to generate such high ionization efficiencies is that ion-molecule reactions occur very readily in an ion source, and they occur particularly well with complex molecules, as might be observed with an organic mass spectrometer. Such reactions results in the formation of larger, less volatile molecules which would rapidly coat the ion source and could also result in sputtering of the inside surface of the ion source and focus lenses. in a gIRMS, we use fairly simple gases, which significantly limits these problems. In part it is necessary to use simple gases to avoid such problems, but also they are necessary to limit the isotopic combinations which could contribute to the observed signal.
A gIRMS system is dedicated to observing multiple ion beams from a specific set of isotopomers. For example, with CO2, we observe the ion beams with mass to charge ratios (m/z) 44, 45, and 46. Inorder to maximize the ion count, a gIRMS uses multiple collectors to simultaneously monitor all of these beams. Effectively, the duty cycle (percent of time spent actually measuring a specific m/z ion) is close to 100% with a gIRMS, whereas it might be less than 1% for a traditional scanning organic mass spectrometer. A gIRMS does have scanning capability but cannot compete with an organic mass spectrometer for structural characterization of compounds.
Because of the high ionization efficiency, the ion beams observed with a gIRMS have currents that are typically in the low nanoamp range (nA). Beams are typically about 1 to 50 nA. This contrasts with organic mass spectrometers which will have ion beams in the picoamp range and below. For organic mass spectrometers, it is necessary to use an electron multiplier for an additional 105 to 106 amplification of the signal from the ion beam. Electron multipliers are not used for the standard gIRMS systems as the beam intensities we operate with would quickly the electron multiplier. In addition, electron multipliers are not sufficiently stable to obtain the desired measurement precision. Instead, a gIRMS system uses multiple faraday cups, which are very stable can can handle the high beam intensities.
Because of the statistical requirement for high beam intensities for high precision isotope ratio measurements, a gIRMS requires nanomole amounts of the element of interest for a good measurement. Typical limitations on an elemental analyzer, for example, are for about 10 micrograms of carbon and about 30 micrograms of nitrogen in the sample). With an organic mass spectrometer, the high amplification of the electron multiplier, along with the need to only see a few ions to obtain a satisfactory mass spectrum, limits the required amount of analyte to about the femtomole scale (10-15 moles of material). There are even works where compounds have been identified and/or quantified on the atomole scale (10-18 moles of material)!
II] Measuring isotopes:
The analysis gases used for C, H, N, O, and S stable isotope measurements are limited to: CO2, CO, N2, N2O, H2, SO2, and SF6. The more different elements are in the molecule, the more difficult it is to resolve different elemental isotopes from each other. Hence, the ideal gas for measuring isotope ratios would consist of single atoms of the element. Unfortunately, it is rather difficult to generate significant amounts of C+ (for example) so we are stuck with the next best case of using CO2 or CO. When measuring isotope ratios it is a good idea for all users to think about what exactly is being measured and how the final isotope ratio is calculated:
For CO2, we monitor m/z 44, 45, and 46. Below is an explanation of the isotopomers and their relative abundances in parentheses.
m/z 44: 12C16O16O (0.989)
m/z 45: 13C16O16O (0.011) + 12C17O16O (0.0004) + 12C16O17O (0.0004)
m/z 46: 12C18O16O (0.002) + 12C16O18O (0.002) + 13C17O16O (0.011 x 0.0004 = 4.4x10-6) + 13C16O17O (0.011 x 0.0004 = 4.4x10-6) + 12C17O17O (0.0004 x 0.0004 = 1.6x10-7)
These indicate that the intensity of the m/z 45 ion beam relative to the m/z 44 ion beam should be about 1.2% (=0.0118/0.989) and of the m/z 46 ion beam relative to the m/z 44 ion beam should be 0.4% (=0.00400896/0.989). The question that the reader should be asking is "how do you get a 13C/12C ratio from these data since you also have 17O isotopes contributing to the m/z 45 signal?"
The answer is a bit complicated, but here we go for a "light" response:
First, all of the ions containing multiple minor isotopes have very low abundances (10-6 and 10-7) so their contribution to the m/z 46 signal is considered negligible. This is not the case for the 17O isotopes which make up about 7.3% (=0.0008/0.011) of the m/z 45 signal. In order to determine how much of the m/z 45 signal is due to 13C, we need to know how much of it is due to 17O. This is done by assuming that the ratio of 18O to 17O is constant in all mass-dependent processes that result in isotope fractionation. Isodat (instrument software we use) lets the user choose between two different methods for this 17O correction. This estimate works quite well in general, but there are some processes in nature which are known to be mass-independent, in which case it would not apply. The correction is made by assuming that the m/z 46 signal is entirely due to 18O. The expected amount of 17O contributing to the m/z 45 signal is then subtracted from the m/z 45 signal by assuming a constant ratio of 18O to 17O. The remaining signal for m/z 45 is assumed to be due to 13C contributions only.