Flame ionization and thermal conductivity have been the most common detectors in gas chromatography (GC) (link to GC page) for decades. Both are compatible with capillary and standard-bore GC. But what makes them tick, and when should you choose one over the other? Let's put them face-to-face to compare strengths, limitations and best applications of flame ionization detection (FID) and thermal conductivity detection (TCD) (links to product pages).
Flame Ionization Detection
First developed in the 1950s1, FID is the most common method used with all types of GC (link to previous blog post).
In a typical FID design, the mobile phase passes from the GC column into an oven, keeping the sample in gas phase. As it moves into the detector, the eluent is mixed with combustion gas, then oxidant. The mixture travels up to a jet nozzle and into the flame to burn.
Voltage applied between the nozzle (positive) and the collector plates (negative) accelerates any reduced carbon ions toward the plates, for detection by a sensitive ammeter. The signal is converted to voltage, amplified, filtered, and recorded by the GC data system. The effluent—mostly carbon dioxide and water—vents through an exhaust port.
Reasons to Choose FID
FID detector electronics filter out high-frequency background noise, for very robust signal-to-noise performance.
Good response to most analytes
Outstanding sensitivity and resolution: down to a few picograms of carbon per second
Wide linear range; generally better linearity than TCD
Simple to use
When Not to Use FID
For analytes that lack carbon-hydrogen bonds (not detectable by FID)
When you want your sample back - destructive method
Situations where volatile combustion gas and flame jet are problematic
For stationary phases that may bleed contaminants at high temperature, or samples high in chlorinated or other compounds that burn inefficiently
Based on technology that predates even FID1, TCD relies on differences in thermal conductivity between target analytes and a carrier gas. Many compounds have relatively low thermal conductivity in the gas phase, so they can produce negative peaks against the background from a highly thermally-conductive carrier such as helium or hydrogen. Other carriers are used as well—in theory, anything that isn't an analyte of interest.
A typical TCD design directs carrier gas flow to two temperature-controlled chambers: one that receives the column eluent (column flow) and another that receives only carrier gas (reference flow). Each chamber surrounds one of four filaments arranged in a bridge circuit.
Electric current applied across the filaments heats them. Carrier gas flowing through each chamber removes that heat at a constant rate, creating a stable background temperature. When sample compounds with relatively lower thermal conductivity pass through the column flow chamber, they remove less heat. The resulting transient increase in temperature changes the filament's resistance, and is detected as a difference from the reference chamber baseline.
TCD detects any species that differs from the carrier gas in thermal conductivity – essentially any gas, given the right carrier. It responds similarly to set analyte concentrations within a class, for example, across organic compounds, so can be used to estimate relative concentration of each peak (ratio of peaks).
Universal; responds to any species that isn't the carrier gas