Microscope use in brewing
For many home brewers the ultimate brewing gadget is a microscope. What could be more more geeky than actually looking at the organisms that make our beer. This article gives interested home brewers guidance on buying a microscope and necessary accessories. It also shows how to make effective use of a microscope.
What to expect from using a microscope
Before we get started, here are some points about what to expect and what not to expect from using a microscope.
The primary use of a microscope in the brewery will be for counting yeast cells using a hemocytometer. Cell counts are useful for determining pitching rate, remaining cell density when bottling beer and determining yeast densities in yeast sediment. The latter is useful for establishing a correlation between slurry weight and yeast count to allow more precise pitching without having to perform lengthy counts.
Yeast counting takes time and easily adds 20-30 min to your brew day. Especially if you want to determine the amount of available yeast before pitching. This is why pitching yeast by weight is more practical if the yeast density can be predicted fairly accurately.
With a simple staining technique (methylene Blue) you'll be able to asses the health of a yeast culture.
A microscope also allows you to better classify the source of haze. In particular it allows you to determine if a haze is caused by yeast cells or much smaller protein globules.
With a microscope you will not be able to reliably detect a infection before it is noticeable in the beer. This is because the bacteria concentration necessary for flavor impact is very low and only by chance you'll be able to see a bacterium in a yeast or beer sample. The use of agar with selective growth media is better suited for this task.
Purchasing a Microscope
The hardest part of getting a microscope is deciding what to get. Prices range from under $20 for kids toys all the way to lab grade equipment in the thousands and tens of thousands dollar range.
For brewing use we are looking for a compound (multiple lenses) light microscope with these features:
A microscope is not the only equipment needed for brewing use. The following is a shopping list of additional items
I buy my lab supplies at cynmar.com but there are many other places on the web that sell this equipment.
Using the microscope
preparing the sample
Since cells are counted only in a very small sample, the sample needs to be representative of the beer or liquid that contains the yeast population. That means that the whole volume needs to be mixed very well and yeast cells cannot be allowed to flocculate. This aspect is very important.
There are two approaches to determining pitching rate, i.e. initial cell density in the beer. Cells can be counted once they have been pitched into the beer and cells can be counted while they are suspended in the propagation vessel. The former has the advantage that it does not require dilution. But if more cells than desired have been pitched they cannot be removed. The later requires dilution of the sample. If the amount of wort added to the yeast is about 5% of the wort volume to be pitched (1 qt for 5 gal batches or 1 l for 20 l batches) a 1:20 dilution is recommended.
For that add 19 ml water to a test tube or other small vessel and then add 1 ml of the stirred yeast culture. Mix well. Don't shake it vigorously. Some yeast strains, in particular ale yeast, like to aggregate in the foam. Repeatedly pull a sample into the pipette and push it back out to flush the pipette.
Poor flocculators like dusty lager yeast and German ale yeasts are easy to deal with. Most other flocculating yest strains will need some help. Here are some practical ways of un-flocculating yest for cell counting 
Maltose inhibits flocculation, which is why yeast un-flocculates in fresh wort. This is also true for heavy flocculators like English Ale yeast (WLP 002). Simply add fresh wort to the yeast sediment and place it on the stir plate for a few minutes. This is a very practical way of determining the yeast count in yeast sediment before pitching since the health of the yeast is not affected and un-flocculating the yeast also allows for a more even distribution when the yeast is pitched.
When the yeast sample to be counted is not intended for pitching, the yeast can be un-floculated with the addition of some sulfuric acid (H2SO4). This does not affect the yeast viability which is important if methylene blue will be used for viability assessment. However, sulfuric acid is highly corrosive and has to be handled with care
Safer than sulfuric acid is the use of EDTA. EDTA (Ethylenediaminetetraacetic acid) is a chelating agent that chelates (“captures”) the calcium ions needed for flocculation. As a result the yeast sample un-flocculates. It does not affect the yeast health and methylene blue staining can be used to asses yeast viability.
Five Star's Powdered Brewery Wash also contains chealating agents and is able to prevent yeast from flocculating. While it is safe to handle and readily available for many brewers it does affect the yeast health and methylene blue staining cannot be used to asses the viability of the culture.
preparing the hemocytometer
Prepare the hemocytometer according to the manufacturers instructions. In most cases this simply means placing the cover slip over the counting grid(s), supported by two ridges.
Now pull a sample with the pipette and create a small drop on the pipette's tip. Place that drop right next to the edge of the cover slip such that it will be sucked into the counting chamber by capillary action. If your hemocytometer has two counting grids, pull another sample and repeat the procedure for the other counting grid. The sample should cover the counting grid while avoiding overflow into the “moat” or overflow area. Too much liquid in the moat may push the cover slip upward and change the volume above the counting grid.
Now place he hemocytometer onto the microscope’s stage and select the weakest magnification. Using the focusing knob move the stage closest to the objective and while looking through the eyepiece move the stage away from the objective until the counting grid can be seen. Focus and change to the next higher magnification level. Re-focus and adjust the stage's position such that the center counting grid is visible. Note how evenly the cells are distributed. If they are clumped together, the sample may need to be re-mixed or the yeast needs more time to un-flocculate.
Unless the focus needed to change, it can be left as is and the next time you count cells you don't have to go through the outlined focusing procedure.
Change to 400x magnification. This is the best magnification for counting yeast cells. By convention the cells in the 4 corner 4x4 grids and in the center 4x4 grid are generally counted (see Figure 13 for the complete Neubauer counting grid). If the yeast density is low, additional grids may need to be included. See figure 4 on how to count cells in one of the 4x4 grids.
When using a hemocytometer with a depth of 0.1 mm (most common) the cell density of the original sample in Million cells per ml (or Billon cells per liter) is
Dilution factor is the number of sample volumes of water used for dilution plus one. Examples:
The formula also means that counting one row of 4 small squares in an undiluted sample gives a very rough estimate of the cell density.
The number of cells in the culture in Billion can be calculated by multiplying the cell density with the culture or beer volume in liter ( 1 Million/ml = 1 Billion/l). Note that this is the volume in which the yeast population is currently suspended.
Cell counts determined with a hemocytometer have accuracy limits that every brewer should understand. The error can be minimized by
The following formula can be used to estimate the statistical error based on the number of cells that were counted:
Ideally multiple samples should be taken and counted separately. If that is not done the two counting chambers should at least be filled with separate pipette loads.
Methylene blue staining
Methylene blue staining allows for a simple way to asses the health of a yeast culture. In theory dead cells will stain blue while living cells remain colorless. In practice it has the tendency to overestimate the viability. Despite its shortcomings, however, methylene blue staining is a quasi standard in the brewing brewing industry for viability testing due to its simplicity and quick results. It is also a convenient test for home brewers since methylene blue is readily available on the internet and has a long shelf life.
In its oxidized form, methylene blue is blue. In its reduced form, called leucomethylene blue, it is color less. This can be demonstrated in the “blue bottle experiment” where methylene blue looses its color due to the oxidation of glucose and reduction of methylene blue. When oxygen is added by shaking, leucomethylene blue is oxidised back to methylene blue and the blue color reappears.
Because of methylene blue's overestimation of viability, results less than 85-90% should be seen as inaccurate when using this stain . For practical brewing purposes if a yeast culture is showing a viability of less than 90%, when tested with Methylene blue, it should not be used and a new culture should be grown. That can be done by propagating a small sample of the old yeast.
To perform a viability test with methylene blue brewers should follow this protocol:
In order to correctly see which cells are stained focus may have to be shifted slightly up or down. Count colorless and light blue/greenish cells as viable and count dark blue cells as dead. Don't count blue stained cell buds if the mother cells did not stain. Buds are busy with growing metabolism and may not reduce the dye. 
cell counts using ImageJ
Cell counts at the microscope take a lot of time and are rather tiring. An alternative is to take picture(s) and count the cells using desktop counting software like ImageJ. Pictures can be taken with a dedicated camera mounted to the microscope or even a cell phone camera that is held close to the eyepiece. The latter will require some practice but works sufficiently well. When taking pictures for cell counting it is also possible to take them at reduced magnification (100x) and thus wider field of view. In this case the only one image has to be taken for each side of the hemocytomer.
ImageJ has both automated and manual counting modes. For the automated cell counting, which is based on particle recognition in the image one needs a fairly evenly lit and high quality image which I'm not able to take with a cell phone. Because of that I'm using the cell counter under Plugins->Analyze->Cell Counter. After initializing the image a cell type is chosen and every click on the image leaves a marker and increments the counter for that cell type. Different cell types can be used for different grids or counting stained and unstained cells.
assessing yeast health
healthy yeast cells are plump and round while starved yeast cells have a more elongated football like shape. See Figure 6 for such an old yeast culture
ale vs. lager yeast
In general you will not be able to differentiate between different strains of yeast, though some strains have larger cells than others. But there is one difference between ale and lager yeast that can sometimes be observed. Ale yeasts tend to stick together after budding and end up forming small stringed colonies consisting of 5-10 cells (see Figure 9). These colonies are more likely to attach to CO2 bubbles and rise into the kraeusen. Lager yeasts separate after budding and only form groups when they flocculate. These groups are clumps rather than chains if cells.
determining the source of beer haze
When a beer exhibits a haze it makes sense to take a look at the beer under the microscope. The procedure is the same as for cell counts with the exception that the hemocytometer should be cleaned more thoroughly than usual in order to avoid particles that are not coming from the beer. I also suggest filling one of the sides with water so you can compare the beer to water.
Unless it is a strong haze, particles will be rather far apart. But you'll be able to see if the haze is the result of yeast or protein complexes. Protein complexes are much smaller than yeast (~0.5 um compared yeast which is 10 times as large). If the haze is the result of a microbial infection it should be noticeable in the beer's taste. Bacteria is much smaller than yeast and usually shaped like rods. Don't confuse rod shaped crystal's (possibly calcium oxalate monohydrate) with bacteria, though. These rods, which can sometimes be found in the sediment, are about as large as yeast cells (see Figure 11).
Yeast hazes will eventually settle. Protein based hazes are much more stubborn and take many month to settle or require treatment with a fining agent like gelatin.
Gallery of various microscopy images
Improved Neubauer counting grid