Introduction: Make Rochelle Salt

About: Just a guy in a place doing stuff. I have always been interested in science. I was the child my parents were worried about leaving alone with the appliances. They were afraid they would come home and find me …

Rochelle salt is a fascinating easy to grow crystal that exhibits piezoelectric and ferroelectric effects.

This instructable will walk you through making your own Rochelle salt from baking soda and cream of tartar. And also show you how to clean your cream of tartar which probably includes some cornstarch.

I have seen a lot of information some good, but mostly just a recipe to follow which I never find entirely satisfying... Do this because I said so doesn't help me understand.

While researching on the internet I could see that the Rochelle salt recipe does not appear to have changed much since it was first transcribed from some book. There was little discussion about theory or properties, beyond it is piezoelectric and this is the recipe.

Showing you how to mix the salt is but one step in exploring the crystals.

From Wikipedia
Potassium sodium tartrate tetrahydrate, also known as Rochelle salt, is a double salt of tartaric acid first prepared (in about 1675) by an apothecary, Pierre Seignette, of La Rochelle, France. Potassium sodium tartrate and monopotassium phosphate were the first materials discovered to exhibit piezoelectricity.[3]


Rochelle salt is deliquescent* so any transducers based on the material deteriorated if stored in damp conditions. It has been used medicinally as a laxative. It has also been used in the process of silvering mirrors.


The starting material is tartar with a minimum tartaric acid content 68 %. This is first dissolved in water or in the mother liquor of a previous batch. It is then saponified with hot caustic soda to pH 8, decolorized with activated charcoal, and chemically purified before being filtered.

So the highlighted bit says Rochelle salt is made industrially by dissolving cream of tartar into solution and then neutralizing with Caustic soda (Sodium hydroxide) to pH 8.

This isn't the common method on YouTube** and Instructables

So I started searching, knowing there is usually more than one set of precursors. I ran across a couple old indirect references, like a discussion around the turn of the century regarding alum based baking powders vs Rochelle salt based powders... It hadn't occur to me till reading it, that is exactly what a sodium bicarbonate and tartaric acid baking powder would make. Modern non alum baking powders may use other recipes, like mono-calcium phosphate.

So what I have found (again) is this, an example of a "lost" technique. In that 1 version of this recipe is being copied and spammed as if it were the only way to do a thing. Were I trained in chemistry I might have immediately seen this.

The internet doesn't know as much as books, or actual chemists, about some things still. While the typical method of converting to washing soda does lend itself to lab techniques better than store bought baking soda... nobody is measuring the weights for it to matter.

A very very thorough blog about Rochelle salt and its history also using this recipe also the place I ran across cleaning the cream of tartar

** a number of YouTube videos these days are about some free energy nonsense using Rochelle salt and/or Epsom salt to make a simple electro-chemical cell and thinking they invented something over-unity. If you are using dissimilar metals you made a battery. Crystals that precipitate in water still contain water making them a rather concentrated electrolyte. In this case Rochelle salt holds 4 water molecules - tetrahydrate. When you put 2 different metals in contact with this you create a voltaic or galvanic cell. These work like the lemon or potato clock science experiment from school.

Step 1: Chemistry

Rochelle salt can be easily produced through at least these reactions;

KC4H5O6+NaHCO3=KNaC4H4O6+CO2+H2O (cream of tartar + baking soda = Rochelle salt + carbon dioxide + water)


(2)KC4H5O6+Na2CO3=(2)KNaC4H4O6+CO2+H2O (2 cream of tartar + washing soda = 2 Rochelle salt + carbon dioxide + water)


KC4H5O6+NaOH=KNaC4H4O6+H2O (cream of tartar + Lye = Rochelle salt + water)

Traditionally you mix the tartaric acid in near boiling water to get as much of it as possible to dissolve. Then mix the alkali in slowly.

Adding baking soda to hot liquid is not very exciting, adding lye to hot liquid on the other hand is. Don't do this quickly

Washing soda and lye are caustic and irritating, they also heat up when added to water. You could have a potential steam explosion like event if you dump lye powder into too hot an acidic liquid. The overall reaction has a negative energy it requires heat from the environment.

The enthalpy of solution (fancy way of saying how much the temp changes when dissolving stuff) for the 3 given alkali reactants are

Lye -44KJ/m This releases 44 Kilo-joules/mole when added to water if we had 100g of water the temp would go up ~6 degrees °C

Washing soda -28KJ/m this releases 28KJ... this temp would go up ~2 °C in 100g of water

While baking soda at 17KJ/m absorbs energy from the water to dissolve. This means the temp drops slightly.

Also of note Rochelle salt has a positive enthalpy like baking soda. I was unable to find this number, but know the solution got colder when dissolving the salt.

Rochelle salt melts at 80°C-100°C (176°F) *sources have some disagreement, but its below the boiling point of water.

Solubility charts are quite informative, the chart I included is the data I found on solubility, and hre is another chart. this whole chapter is a decent read, it also includes a diagram of a scientific crystallizer that's in use today.

At 25°C you can dissolve equal weights salt and water, at 45°C you can dissolve 200% of the weight in salt in water.

Fractional crystallization and washing first with ice cold water then with a non polar solvent like isopropyl alcohol have yielded me slightly cleaner results, but nothing beats pure reagents before it becomes this super water soluble salt.

Step 2: Gather

For the % of us still using imperial measurements. I will try to include Fahrenheit degrees where I remember but you can do the ounce/fluid ounce conversion yourself. It just makes chemistry harder than it needs to be to also add a unit conversion.

Glassware - 2x something large enough to hold twice the volume of your recipe.

A stirring stick - ideally glass or non reactive plastic, Stainless steel is probably ok

A gram capable scale - the range will depend a little on your batch size You can do this without a scale but its kinda irritating

coffee filters - or actual filter paper if your inclined/lucky

A hotplate or stove

A food thermometer (something that will measure up to boiling. a meat thermometer could work in a pinch)

Stove safe pot to use as a double boiler for your Glassware

(Recommended) purifying your cream of tartar you will also need

A .5L (2cup) microwave safe measuring cup

A microwave

A large container to hold the cooling tartar solution

Some space in a fridge or other cool area < 10c (50f)

A large flat glass dish for drying (pyrex cake pan)


1.Cream of tartar aka potassium bitartarate - potassium hydrogen tartrate -

2. Baking soda aka sodium bicarbonate - sodium hydrogen carbonate - bicarbonate of soda

This reaction can also be done using washing soda, which most instructions will have you make from baking soda by cooking. The only advantage to washing soda over baking soda is you know there is no moisture in the weight when you bake it fresh.

This can also be done with Sodium Hydroxide - commonly called lye, a caustic somewhat hard to find and mildly dangerous chemical.

3. Distilled Water - some may be able to use tap water but this is generally just considered a nono when dealing with chemistry. Do you have fluoride or ammonium chloride in your tap water what about dissolved iron, copper or minerals? All of these will interfere to some degree and create unwanted byproducts.

Step 3: Clean Food Grade Chemicals

Cream of tartar is almost insoluble in water at 0.37g per 100ml at 20c and a whopping 6.1g at 100c compared to the salt we are making which can dissolve 100g of its weight at 25c.
Knowing this we can get pretty pure cream of tartar using the following technique which employs fractional crystallization. When you have many things dissolved in water, the least soluble tends to precipitate out first.

Also most things are more soluble in hot water than cold. This comes in handy all through making crystals.

If we add enough water to completely dissolve all the cream of tartar and then pour off the water. Anything that did not dissolve is insoluble in water and not the product we are after. One of my bottles of tartar had a much higher concentration of the even less soluble form of tartar. I had to add another 400ml of water to dissolve it all. you can also re use the water for purifying after it has cooled and most of the tartar has dropped back out.

Once the water is cooled most of the cream of tartar precipitates out while most impurities will stay in solution.

This is based on a 100g container of cream of tartar, as long as you keep the water to powder ratio about the same for this step it will work. With this volume of water depending on how low a temp you go, you are losing 2g or less of the tartar per 25g going in. Anything else lost is going to be impurities.

I modified this recipe from 600ml for ever 37g of cream of tartar.

Purify cream of Tartar (credit for this step goes to M. Gasperi)

  1. 25g Cream of tartar + 450ml distilled water in a Pyrex container
  2. Bring to a rapid boil in a microwave oven
  3. Let stand 1 minute and then stir
  4. Bring to a second boil
  5. Let stand 1 minute and then stir
  6. Bring to a third boil
  7. Let stand 1 minute but do not disturb it this time
  8. Gently pour off just the liquid into a large bowl leaving any solid residue
  9. Repeat step 1 adding to the residue already in the container
  10. Repeat steps 2 to 8 but you can now discard all the residue
  11. Allow the bowl of combined liquid to come to room temperature
  12. Refrigerate the liquid for at least 10 hours down to at least 10C
  13. Occasionally stir just to knock down any crystals floating on the surface
  14. Gently discard just the liquid
  15. Keep the residue which is pure Cream of tartar
  16. Completely air dry, any moisture will throw off your weight measurements

The pure Cream of tartar forms as tiny beautiful glittery crystals. The amount you recover depends partly on how good your tartar was to start with. I typically lose about 12% by weight since contaminates are thrown out in the solid residue and also dissolved in the discarded liquid. Starting with 74g you might only end up with something like 60g of pure dry Cream of tartar, but that is still enough to make about 90g of Rochelle salt.

I think most baking soda is pretty pure since bicarbonate is cheap. You shouldn't need to clean your baking soda, but if your soda yellows when baked to washing soda then you should get a different brand or clean it thusly.

Purify Baking soda

I didn't think this (cleaning baking soda) was really necessary until I started making pound sized batches and found a very small amount of cornstarch or some other flow management in the baking soda. after reduction I have a amber sweet smelling liquid, that reminds me of the caramel corn shop. Great smell bad for crytals.

236 g/L (100 °C) of baking soda can be dissolved in water.

Use the recipe and process for cream of tartar above substituting 100g of baking soda + 500ml distilled water

To make the baking soda into washing soda simply cook above 60C (140f) Reading suggests temps in the range of 150-200c (300-400f) are quite acceptable for speed of conversion. Its not something you can really over cook.

  1. place baking soda/slurry in a pyrex cake pan or other oven proof glass flat container more surface area is better
  2. put in oven and turn on to warm/low
  3. cook at low till dry
  4. once dry, turn heat to 175c(350f)
  5. cook for an hour or 3 and stir once in awhile to make sure heat, atmosphere exposure gets to everything, the soda will change to a shiny pearly white when it is done.

Step 4: Practical

Washing 100g Cream of tartar yielded me 88g pure crystal with all the water I used and my poor handling this means my cream of tarter is actually fairly pure. But in testing I found out I could tell a purity difference the reaction with the unwashed taking longer and the resulting solution has a noticeable tan/yellow tint. The uncleaned tartar produced cloudy tan crystals without a lot of large monolithic growths.

To be properly usable this needs to be cleaned, and with Rochelle salts love for dissolving in water this is now not as easy as it was at the cream of tartar stage.

The cleaned solution was crystal clear and the resulting crystals are clear white which is essential if you were planning on using any of the optical or electrical characteristics of these crystals.

  • Measure your pure dry cream of tartar weight
  • multiply tartar by .45 and measure out this weight in baking soda (because of water wieght you may need more of this than calculated)
  • Or .21 for NaOH or .28 for Washing soda

  • multiply tartar weight by 1.3 and measure this out in water weight
  • For me this was 88g cream of tartar 110g water and 39.6g baking soda

  • in a 500ml (2 cup) pyrex container place cream of tartar (a 250ml (1 cup) will work but it fizzes up so you might lose some
  • add water

  • place on the double boiler and bring the water to 70c

  • once at 70c start adding baking soda, mix in small amounts slowly
  • wait for the bigger reaction to subside and repeat
  • cook at 70c while mixing till the solution is clear and there is nothing at the bottom
  • once the solution is clear pour through a filter **
  • set someplace to cool slowly.

SAFETY TIP: when filtering do not completely seal the edge, esp when using hot fluid, the air in the container has to go someplace and expanding air will spill your hard work.

If using a large 2 cup container you can mix in almost 1/4 of the bicarbonate at a time. Till the last 1/4. Even though you weighed things from here add in small amounts at a time and be sure it still causes a reaction.

*This took about 40 min for the cleaned tartar and 50 for the uncleaned which also had a little too much baking soda in the end.

**I filtered the solute then placed it back into the pot of the now turned off double boiler and let the whole thing cool slowly overnight.

Step 5: Growth

Once everything is dissolved filter into a clean container and let cool,

If your solution is not water clear you still have impurities, it also may appear a little strange since it has a higher refractive index and is light polarizing. It should not have any yellow.To purify further simply grow crystals. Remove from the liquor, the crystals are cleaner that the solution.

When you have a decent layer of crystal growth stir the liquor and decant through a filter. If the solution is fairly pure I air dry the crystals in a tray at a slant so water can run off the crystals and dry away from them. If the solution is rather dirty, I will also wash the crystals with as little distilled water or clean liquor as possible.

After cooling the first batch, If you pull out 1 or 2 large good crystals and reheat the solution with the remaining crystal mass you can re-cool for more crystals. I suggest stirring and filtering.

You can do this a couple times until the salt:water ratio is too far from ideal, either you don't have enough salt to make useful amounts or your heating has driven off too much water from the batch and you get freezing transitions rather than crystallizing.

Simply put, water can only dissolve so much salt at a given temperature, or another way to say it is the maximum salt to water ratio is determined by temperature.

That simple thought has a lot of complex reasons and consequences.

As the temp increases more salt can be dissolved, as the temp drops the water has to let go of any excess salt.

If we remove water, either chemically or through evaporation the salt ratio increases, and solids precipitate out in response to equalize.

The practical application; if too many crystals are forming, or crystals are too foggy either lower concentration of salts, or cool even slower. Conversely, if crystal growth is pathetic try more concentrated solution. Crystal growing requires patience, but if a super saturated solution isn't growing crystals after 2 days it may need seeding.

Inorganic crystal growth tends to have a delay between supersaturation and spontaneous nucleation. Similar to super cooled water freezing when being poured out, the system can be outside of stable zones but not have enough energy to trigger. In the example above of no crystals after 2 days I have not managed to get a seed in without causing the rest of the solution to precipitate out excess.

The crystal for the cover photo was made by placing a seed in a just down to room temp super saturated solution. This makes it easy to make large somewhat poor crystals. There are too many inclusions and voids for it to be good for much more than sitting o a shelf.

Evaporation being a slow process allows time for the crystal to assemble well and is generally better at growing large crystals for those with patience. Dust free temp control environments are helpful. basements root cellars and other low temp swing environments.

The crystal used for the cover shot was grown by dissolving ~365g of rochelle salt crystal into ~250ml of water and then placed into an insulated cooler full of hot water. The huge mass of the water in the cooler and the insulation will slow down heat loss allowing for larger crystal growth to be possible.

With those amounts I expected saturation to occur around 36c and ~150-250g of crystal to precipitate out.

I determined this by using the solubility curve of the salt. I've seen a lot of these curves and they tend to vary for this salt by quite a bit. I think this may be related to the chiral nature of this crystal and differing solubilities of the enantiomers. Rochelle salt's solubility is pretty good, and rapidly increases as it heads towards its melting point which is around 75c.

sodium chloride (table salt) is an easy salt to determine solubility with its rapid crystallization and flat curve.

Just for a touch on the complexity, I wont really try to explain everything here, first I don't know it myself, and second there is a whole scientific discipline I (a layman) am trying to condense into a couple paragraphs...

Growing a crystal has been likened to a complex game of tetris. Micro crystals exist in solution and in proper circumstances will attach to a larger crystal surface. Unlike tetris this isn't a 1 way thing, there are plenty of orientations and a particular micro crystal could attach and detach many times looking for a proper fit. Micro environmental changes influence bonding. Most changes are resisted so there is a battle between many factors Example; by attaching to the bulk surface the local solution is now deficit and can hold more salt so the local area is now trying to dissolve the crystal, while diffusion from the surrounding area tries compensate. Simultaneously the crystallization releases heat which further decreases the effective salt ratio. The crystal is trying to form in an ideal structure to equalize energy micro states - something about the 3rd law of thermodynamics

Water is amphoteric, and does these complex hydrogen bonding things, its basically liquid magic. Its neutral, acid, and base simultaneously. ie H2O + H2O⇌(H3O+) + (OH−)

When we dissolve salts in solution we are trying to achieve chemical equilibrium the anion and cation disassociate

In un-stirred solutions you have diffusion and thermal variations ie there will be more salt at the bottom of the water column than the top. Unless you are using the evaporative method where the surface probably has the highest concentration.


Some of the important factors influencing solubility are:




Ionic Strength:

As you can see even a brief toe dip into crystal growth kinematics is a trip down a deep rabbit hole. Luckily we don't actually need to know all this to get large crystals.

The 2 main methods for growing crystals at home are cooling, and evaporation.

Cooling is good for seed crystals and rapid crystal growth.The slower you allow something to cool the larger a crystal you can grow within certain parameters. Induction period and metastable growth zone are easiest to affect by temp and concentration. eg Higher concentrations nucleate faster, higher temperature tends to increase the speed crystals can grow at to a point. too high a temp decreases growth rate and increases bulk defects. The commercial processes seem to focus on temps between 33-38c for Rochelle salt.

Step 6: Links and Reading

In order to do anything more than use Rochelle salt crystals as a laxative (its original design purpose), or a lab chemical you need to identify the active planes of the crystal. On 1 axis its electrically active.

The axis terminology for piezo crystals seems to normally be, either XYZ (planar) or when referring to cuts 000-111 (miller indices)

X is the electrical axis face to face across the center - Electrical input = mechanical deformation and polarization change This is the basis for piezo microphones, speakers, pressure sensors, and timing crystals*. These are transducers, to change 1 form of energy to another.

Y is the mechanical axis corner to corner across the center - Mechanical input = electrical output This is how a piezo lighter works, although it uses quartz which can withstand the tremendous shock required to generate 5-10k volts

Z is the optical axis 90 degrees from the 2 previous planes - many piezo crystals can be used as polarizing devices or frequency doubling with the correct cut and strong enough beam. A level of photon saturation is required to create the cascade that leads to second or third harmonically generated frequencies. In Rochelle salt this is the easy axis to identify. The crystal tends to grow much faster on the Z axis and If you didn't check really carefully you wouldn't notice the Z faces are not actually parallel

All links are the property of the respective owners and were merely found via google search and while every crystal system reacts a little differently the concepts from 1 usually apply to another.

A more detailed description of the crystal axes*

MIT tells you how to grow crystals in simple terms -

A very thorough blog on Rochelle salt and probably the source of a bit of my inspiration for sharing -

Handbook of piezoelectric crystals -

The PDF links on are the doctoral thesis from Dr. Christine Strege

I really suggest trying to get a handle on the concepts discussed if you want to understand how to grow better crystals. While coming from the lab setting and assuming an expensive crystallizer the information is still applicable to understanding better ways to get higher quality with your setup. For student projects just temp control via a large thermal mass and evaporation is probably all that can be simply accomplished.

"state of the art" an introduction to the concepts and terms of crystallography if nothing else look at the pictures.

While about Epsom salt reading this and understanding any of it will help you understand some of the complex mechanics of crystal growth and the effects of various influencing factors.

Long and thorough discussing ADP/MAP and crystal inclusions

Step 7: Single Isomeric Production of Rochelle Salt

The obvious trick here is to start with a single isomer precursor. You can buy single isomer tartaric acid.

I've screwed this process up in all kinds of ways and using the ingredients given its nearly impossible to get wrong. You can reduce yield, but you don't get any real byproducts, you just end up with unused precursors.

With large batches using baking soda I have run into the following situation. The addition of more baking soda doesn't react, and there is still large amount of solid in the bottom. Earlier I went over the enthalpy of formation for each reaction, and baking soda requires energy to work, so the reaction gets colder. Somewhere around 10c(50f) the reaction basically stops and more energy is required. I use a slow cooker to mix large batches in now.

This will only be as isomerically pure as the precursor chemical l-(+)-tartaric or l-(r,r)-tartaric. unlisted tartaric will still be good fun, and likely contains some of all 3 isomers, which will still be purer and possibly cheaper than food grade with its cornstarch filler.


Dangerous chemicals and real chemistry time. WEAR PPE, extremely caustic chemicals. Gloves and a face shield min, the chemical burns from KOH, or NaOH are bad. Keep a bowl of lightly soaped vinegar around to neutralize skin exposure.

The KEY to preventing serious chemical burns after all the Personal Protection Equipment is very rapid response to exposure. 60 seconds is too long to go before a burn has started.

IF you suddenly think you taste metal or salt, go rinse exposed skin.

IF any place being rinsed that feels like it has soap on it that's the Hydroxide turning your skin into soap, in another 45 seconds its going to start hurting.

ONCE you know where the exposure is add a little vinegar to the area and continue rinsing until the slick soapy feeling is gone. Water alone is usually enough but if you came into contact with the solid hydroxide or drops that melted, vinegar is about the only way to neutralize it quickly enough to prevent chemical burns.

Sodium and Potassium hydroxide chemicals are deliquescent, meaning if you drop a bead or flake somewhere, it will absorb water from the air and be a drop of of caustic liquid by the time you see it again.


For this you will need

Food grade l-Tartaric acid

Potassium Hydroxide

Baking soda

5% White vinegar

100 ml graduated cylinder

1-10ml graduated cylinder

PH meter or test strips or ideally phenolphthalein indicator

accurate scale .01 or better

We start by determining how pure our chemicals are.

we will do this using titration against a known acid solution, household white vinegar!

weigh out 4.66g of KOH or 3.32g of NaOH (I refer to NaOH if you don't want to use baking soda, real lab work)

add this to slowly to 50ml of water in the graduated cylinder, This is very reactive and produces a lot of heat, cold water is best, never point this at your face treat like a gun barrel while dissolving. Once dissolved, make the volume up to 100ml. IF the chemicals were 100% pure this would make a solution of identical strength by volume as 5% white vinegar.

add 1ml white vinegar to a small graduated cylinder, then add 1ml of the hydroxide solution to it. test pH it should be acidic or with perfect chemicals neutral. slowly add more hydroxide solution until the pH become neutral. Don't forget to agitate and rest the solution for a few moments if your not using indicator.

Once you know how much hydroxide was used the purity is determined as 1/ml used. so if it took 1.1ml your product is 1/1.1 or 90.9% pure.

you can dilute both solutions by equal amounts for better accuracy. IE dilute 10:1

My batches

Measure out 151g of L-tartaric acid (1mole)

add ~200-250ml of distilled water stir till mostly dissolved (I use the water from the previous batch which is contaminated with leftover l-tartaric and dissolved cream of tartar.)

Measure out 56g of KOH (1 mole) then using the purity % we came up with earlier measure out the required extra into a seperate container.

add the larger amount to ~70ml COLD water allow to dissolve and cool back down to room temp (I use a water bath).

Add the smaller amount to 10-20ml water

while stirring add the larger volume of KOH slowly to the tartaric solution over 3-5 minutes, the solution will get warm

once added allow to rest a few minutes then decant the liquid into another reaction vessel

The liquid at this stage still has a quantity of free tartaric acid, and a small amount of the nearly insoluble potassium salt (cream of tartar), but because of the buffering nature of the Potassium salt removing the bulk of it is desirable.

Drop wise add the smaller amount of KOH stirring and resting occasionally until there is no longer precipitate forming.

This should get you the closest maximum yield.

allow to settle and decant this liquid. Set aside for the next batch if you like, to dissolve the tartaric acid in.

all of that nice white powder should be Cream of tartar, with a small amount of tartaric acid

A simple wash with 200 ml of ice cold water will remove any leftover tartaric acid or KOH, both of which are far more soluble in water than the cream of tartar.

Adding too much of the KOH starts redissolving the cream of tartar as it starts buffiering the PH of the solution against the now free -OH ions, While free K+ and OH- ions are not a problem to crystal growth(appear to be helpful in some studies of rochelle salts) I want to start with pure and add impurities intentionally rather than just hope whats leftover doesn't interfere.

This is now ready to add the sodium too, to create Rochelle salt, you can use any of the products previously mentioned. Lye, Washing soda or baking soda.

I have tried with lye, washing powder and baking soda, and will say I prefer to use baking soda after many many batches.

It is the simplest to tell when you have added enough. It stops bubbling as soon as you add it, unlike all the other options...

Maker Olympics Contest 2016

Participated in the
Maker Olympics Contest 2016