Selecting a Detector for LC

Although a blog is not a good way to teach all there is to know about LC detectors, I have tried to put together information to give an overview of the most common detectors.  Restek does not sell HPLC instruments or detectors, but we think we might be able to help if you are very new to the world of LC. Before getting started, if you have a textbook on analytical chemistry, it would most likely be useful also to review the section on LC detectors. Since new techniques are developed frequently, to obtain the most current information, i.e., the “latest and greatest”, the information offered online from your instrument manufacturer would be helpful as well.

For convenience, I have summarized and rated some key attributes in the table below for the detection methods discussed in this blog. Please keep in mind that all of these will vary depending on the specific method and analytes, perhaps different detector models and vendors, and also the level of training for the analyst. The attributes listed are rated on a scale of 1 to 4, with 4 being the highest and best, 1 being the lowest.


Detector Selectivity Sensitivity Ease of use Linearity
UV 2 3 4 4
PDA/DAD 2 3 3 4
Fluorescence 3 4 3 2
ELSD 1 2 1 1
RI 1 1 2 1
MS 3 3 4 3
MS/MS 4 4 3 3


UV Detector

The UV Detector is by far the most common detector used for liquid chromatography. Sometimes these detectors are offered as UV-VIS detectors, which can also measure absorbance in the visible range of the electromagnetic spectrum.   Absorption of UV light occurs when the energy applied causes electrons to jump from their ground state to an excited state, an orbital at higher energy, to be exact. Functional groups that containing electrons that easily absorb UV are called chromophores. Often the best chromophores are functional groups that contain conjugated double bonds or delocalized pi electrons, such as in a benzene ring.

To determine the best wavelength to use, generate a wavelength scan and look for the wavelength(s) at which maximum absorbance occurs.   An alternate approach is to identify wavelengths that have the strongest absorbance for specific functional groups in the molecular structure.  Tables are available in literature for this, for example, the “Correlation Table for Ultraviolet Active Functionalities” (Bruno and Svoronos), found in the CRC Handbook of Chemistry and Physics (

Response is determined by Beer’s Law, which demonstrates that absorbance is directly proportional to concentration:


  • A=absorbance, E= molar extinction coefficient (molar absorptivity), L=path length of detector flow cell, c=concentration of analyte (Molarity)
  • E depends on structure and properties of the analyte, as well as wavelength.


Diode Array (DAD) or Photodiode Array (PDA) Detectors

DAD and PDA are one in the same; terminology just varies between different vendors. In a practical sense, these are a 3-dimensional version of a UV detector.   A diffraction grating enables the detector to continuously and rapidly monitor across all wavelengths simultaneously.  The result looks like a chromatogram for UV, done at multiple wavelengths.

A DAD or PDA may be particularly useful in method development or research projects where there is a need to analyze for unknown compounds, possibly with unknown functional groups attached. The advantage is the ability to collect data from any selected wavelength and to change that selection after the initial data workup if desired.

Fluorescence Detector

Fluorescence light is emitted when a higher energy, short wavelength of light is absorbed and as a result, a lower energy, longer wavelength light is emitted. Fluorescence detectors are very sensitive and also very selective. However, since very few compounds have natural fluorescence properties, often derivatization is needed to attach fluorescing chromophores. This may be done either pre or post-column, and usually requires an additional pump and a rotary switching valve. Post-column derivatization is always safer for the HPLC column, although the setup is sometimes more complicated.

Similar to UV, compounds that emit a strong fluorescence signal usually contain conjugated pi-electrons or aromatic rings. The intensity of the signal is determined by the excitation and emission wavelengths. Typically, the excitation wavelength is held constant and the emission wavelength is varied for maximum response.


The ELSD or “Evaporative Light Scattering Detector” is a good alternative to UV if your analysis is for a set of compounds that lack any UV chromophores, when you wish to avoid derivatization, and when an LC/MS or LCMSMS is not available.  The ELSD works by nebulizing the sample and measuring the degree of light scattering, which is related to the mass of the analyte.  The ELSD is used very commonly for sugar and for carbohydrate analyses and also sometimes for lipid analyses.  It is somewhat limited in that no buffers can be used in the mobile phase and the detector often exhibits a nonlinear response, or at least it has a very limited range linear response.

Refractive Index (RI) Detector

RI detectors can be universally applied to a wide variety of analytes, since no chromophores are required.  The detector works by measuring the refraction of light through the sample as it passes through flow cell.  The refraction from the sample is compared against the background produced by measuring mobile phase alone in a reference cell.   Similar to ELSD, this technique might be used when mass spec is not available. It is also usually less expensive than an ELSD.  Drawbacks include possible issues with interferences, low sensitivity, and the inability to use with gradients or buffers.

Mass Spectrometry (LCMSD or LCMSMS)

Although the most costly technique in terms of capital expense, mass spec has quickly become one of the primary detection techniques for LC.  With LCMSD, the mass-to-charge ratio of ions, shown as m/z, is measured, which can be very selective.  LCMSMS is even more selective, since it measures the mass-to- charge ratio of parent (precursor) and daughter (product) ions after subsequent fragmentation.   One advantage of such selectivity is that analytes can usually be distinguished from one another, even if they coelute.  This allows for more compounds to be analyzed simultaneously, which particularly useful for screening methods.

LCMS works by treating the sample/analyte molecule as follows:

  • IONIZE- Occurs at the source. In positive mode (most common), protons are attached. In negative mode, electrons are attached/ protons removed from precursor molecule. (precursor=parent). The most common sources are Electrospray and APCI (chemical ionization).
  • FOCUS- Ions are drawn through an orifice and skimmers by vacuum, guided by the octopoles or ion bridge into mass analyzer, with assistance of nitrogen as both a drying gas and as curtain gas.
  • SEPARATE- The separation step is done by a mass analyzer or mass filter. Quadrupole, Time of Flight, and Ion Trap are the most common.
  • DETECT- Signal generated when ions reach the detector and trigger release of electrons. Example: electron multiplier.

LCMSMS works the same way, except the analyte ion undergoes fragmentation after the initial separation, usually in a something called a Collision Cell. The collision cell is located between the two mass analyzers.  The fragment ion, called a product or daughter ion, is separated in a 2nd mass analyzer. The sequence of events for the overall process with LCMSM can be described as this:


The greatest benefits from LCMSMS are realized by operating in MRM or SRM modes (Multiple Reaction Monitoring or Selected Reaction Monitoring, respectively.) Terminology differs by manufacturer. In this mode, MRM pairs are selectively monitored and reported. The pairs are represented as m/z of the parent ion followed by  “>” and then the m/z of the fragment or daughter ion.  A good example is for trans-3-Hydroxycotinine, from this chromatogram:

In the above, trans-3-Hydroxycotinine is identified by MRM pair m/z = 193.1>80.0, and also a second MRM pair is m/z=193.1>134.0.

These are just a few details about LCMSMS that I have highlighted.  If interested in learning more, I suggest contacting an instrument manufacturer and/or enrolling in a course for training.

For more information about all of the detectors we have discussed here, below are a few links that you may find useful from various sources.

From Hitachi-High Tech:

From Shimadzu:

From Bioforum:

From Chromatography online (LCGC magazine):


I hope this overview of detection methods has been helpful. Thank you for reading.


Tags: , ,

7 Responses to “Selecting a Detector for LC”

  1. Amit Gujar says:

    Hello Nancy, Thanks for the informative blog post. I don’t necessarily agree MS detector being highest on “ease of use”…

  2. Hello Amit and thank you for your feedback. The ratings I gave are based on a combination of my personal experience and my impression from the written information that is available. Other views are certainly welcome here and we value the observations from your experience. I would also encourage you to share why you might have rated it differently and what you experienced specifically, if you would like. Thank you again for reading the post and participating in the conversation.

  3. Amit Gujar says:

    Well, most of my experience has been with GC and GC/MS systems and from my experience the MS detector needs a lot more attention than say FID …we need to be sure it is tuned and calibrated correctly before injecting the sample…and just the overall care that goes in maintaining as well as data analysis is definitely not something a novice user can do easily from the get-go. My only experience with LC was with the “lowly” RI detector which I thought was the simplest detector ever (but you ranked it at 2) but then again I was doing simple separations of known compounds without any matrix…
    I am curious what other people think about your ease of use rankings. Hopefully we will have more feedback from other people. Thanks once again for posting this.

  4. Thank you again, Amit. I appreciate the details and explanation. I will have to say that the RI detector I used was an older model and it was quite a chore to maintain a stable baseline. In my case, this was related mostly to the mobile phase in the reference cell. I have heard that newer models are much more stable and easier to use, although I have not had the occasion to use one of these. Conversely, my experiences with an LCMSD were applications that were pretty much perfect for the instrumentation and the methods were well established. The analytes were fairly polar herbicides and very easy to ionize. The sample was a plant extract that was fairly well prepped beforehand. The MS portion of the project ran without a glitch throughout and I did not have to stop for maintenance in that regard. In fact, with at least one of these projects that I remember, the chromatography itself required much more attention, due to the use of ion pairing agents.

    I mentioned also that training is a factor which makes the ratings more objective. In my case, I probably had better training on LCMSD than I did on using an RI detector. So.. this is a good example of how perceptions can vary.

  5. Jason Hoisington says:


    I think the fact that you can describe your MS applications as “perfect” and “well established” is a big part of why you have the ease of use so high. In my experience (primarily with GC/MS, as with Amit), running a mass spec with ideal analytes and a simple matrix is a much different experience than touchy analytes in a complicated matrix. And while the same is true of simpler detectors like FID and VWD it’s not to the same extent as you’d see in MS. That brings up robustness as another thing to consider when deciding on a detector, but that’s going to be so sample dependent that it’s hard to give a good idea of what to expect without knowing something about the application in question.

    Another thing that I think is appropriate to consider is how complicated the data interpretation is. MS, and to a lesser extent DAD, give quite a bit more information than just the basic chromatogram you’d get from most other detectors. Again, depending on the complexity of your sample this can make the data much more difficult to correctly interpret. MS/MS can simplify it in some respects due to it’s greater specificity, but the initial set up and selection of ideal mass transitions is a lot more up-front work than you’d need for most other detectors.

  6. Thank you for the input, Jason. I agree that data interpretation is more complicated, but in my situation the LCMSD data was much simpler than what is typical for LCMSMS, at least. I have not done much with GCMS, but it sounds like LCMSD or LCMS may be a little less finicky. I would interested in hearing what others have experienced along these lines.

  7. Amit Gujar says:

    I think if a person starts off his chromatography journey with a analog detector he can find the extra dimension of data that MS adds to be a bit confusing…i can tell that from my own experience…when I started using the MS; for a long time I would quantitate on the TIC (Total Ion Chromatogram); just like I had done in the past for FID (see a analyte peak-integrate it) till somebody pointed out to me that you should be doing it on the extract ion (XIC) for your analyte and that you are not taking the selectivity advantage that the MS provides by integrating on the TIC…

    I too would be interested to hear if MS for LC is more robust and less finicky than GC (or not)…it would be nice to hear from people who have used both GCMS and LCMS extensively and equally…

Leave a Reply

Restek Domestic Customer Service



Your Full Name

Your Email

Company Name


Spam Block (Please leave this blank)

all fields required

Thank you

Your message has been sent. We will be in touch shortly.

Message not sent

Sorry, your message could not be sent at this time. Please try again later, or contact Restek or your local Restek representative via phone.