I'm considering various lab supplies for a protocol we're going to be running, and have been struck by the remarkable difference in price for different quality levels of the same basic substance.

Sigma will sell you water, for example, for anything from \$15/L to \$500/L and even over \$10,000/L, depending on what level of purity is guaranteed. Obviously, one doesn't want to use inferior tools and guaranteed unreliable results, but also it would be foolish to use a reagent that's far pricer than one's actual tolerances.

Most of the actual protocols that I encounter, however, don't seem to actually specify what grade of purity they need, and I'm trying to figure out how to navigate the problem. Trial and error is definitely not a good option, since (a) it can waste a lot of time, and (b) good and bad results in a trial don't actually tell you what you need (a good result could be an unusually good batch of poor reagent; a bad result could be from an entirely different problem). Asking the authors is also not a great solution, for a number of similar reasons.

To make this concrete, let's consider a very simple and typical case that I am currently encountering. The protocol I'm setting up for right now cultures E. coli in LB media. Sigma will sell me LB media powder as "quality 100" L3152 for \$139/kg and "quality 200" L3522 for \$182/kg. Both say they are "good for E. coli".

Is there any reason not to pick the cheaper media? How should one decide how much quality their media needs to have for a typical bacterial culture experiment?


3 Answers 3


It looks like Merck have different levels of quality control for their products. MQ 100 being the lowest up to MQ 600 being the highest. From what I can see the MQ200 product here has defined composition (of salt, tryptone and yeast extract in g/L) whereas the cheaper MQ100 product does not appear to have its composition ensured.

I am sure your cultures will grow in cheaper media but perhaps you would theoretically lose out on some replicability and perhaps if you were doing metabolically focussed experiments you would want to be certain of your media composition.

As for water, I have always used distilled water which is then autoclaved after you add the media with no problems.

  • 1
    $\begingroup$ What quality level do you tend to use in your own experiments? $\endgroup$
    – jakebeal
    Commented Mar 20, 2021 at 18:54
  • 1
    $\begingroup$ @jakebeal I am fairly sure the composition is ensured - I think it would perhaps be best practise. But again, definitely not required for microbial growth or is really likely to have a meaningful effect on your results - I guess scientists can be quite pernickety about these things though! $\endgroup$
    – JEJS
    Commented Mar 20, 2021 at 19:02

The answer depends on the application and level of oversight/compliance you require. Most research labs I've seen use de-ionized or even reverse-osmosis water (from a tap) for most culture media applications. But when I was doing trace-metal analysis, I had to use higher grade water for almost all steps in my experiment. Mostly, we needed a certification with each batch that our specific analytes were below acceptable levels (usually around 1 part per billion). But even "trace-metal" grade reagents often contained iron in the part per million levels, so we had to order even higher grade (pricier) water when we wanted to look at trace-level changes in iron metabolism.

One challenge is that there are few industry-wide quality grading systems. Most of them are specific to a given manufacturer, so it takes a bit of homework to understand what the mean, specifically. For example, it seems like the M-Clarity system used to define "quality" in the link you shared is a Sigma-Millipore specific program designed to help customers with product selection for regulatory compliance. The MQ 100 - MQ 400 levels all appear to follow the same industry manufacturing standards (ISO 9001), and only vary in the level of product analysis and change-notification offered to customers.

If your lab were performing quality control or assurance in a regulated industry, you'd need to ensure that all of your reagents (and laboratory practices) are in compliance with such regulations. If you're in a research lab, these extra features may or may not add any value to your work, depending on the specific applications you have in mind. For most experimental applications, there's not likely much added value in being notified every time a manufacturer gets a different raw materials supplier since most LB ingredients like tryptone and yeast-extract are already complex (undefined) substances. Then again, if your research aim is to optimize growth for industrial fermentation reactors, that information might come in handy.


This answer only concerns Streptococcus pneumoniae.

The lab I worked in used cheap ingredients, the same as everyone else claimed to in this field.

The only shocking thing I found was that the sterile media (already prepared) that I was given to use for DNA extraction did not work (DNA would be fragmented). It turned out the media had been waiting in the incubator for a year. Made fresh media and all was well.

Todd-Hewitt Broth (THB) with 0.5% Yeast Extract (YE)

3.0 % (w/v) of Bacto™ Todd Hewitt broth powder (Difco cat# 249240)

0.5 % (w/v) of Bacto™ yeast extract (Difco cat# 288620)


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .