Real Nanoparticle Ferrofluid From Commonly-Available Materials

In the world of chemistry and material science, few substances captivate the imagination quite like ferrofluid. This mesmerizing liquid, renowned for its ability to defy gravity and morph its topology in response to magnetic fields, has fascinated scientists, artists, and enthusiasts alike since the 1960s. Its unique properties make it a versatile tool in various applications, ranging from cutting-edge technology to captivating works of art.

Unlike many of the other "ferrofluid" tutorials around the internet, over the past two or so years, I've worked on a procedure to create real ferrofluid - nanoparticles of surfactant-coated magnetite suspended in a nonpolar medium - that could be made by almost anyone, and without the need of purchasing expensive specialized tools or uncommon chemicals. The procedure is now at a point where all that can be achieved; small batches of high-quality, non-degrading ferrofluid can be made from basic lab equipment and reactants that can be readily found online or in most local hardware and grocery stores.

The supplies I list here are ideal, but not all necessary. Many can be replaced by cheaper and easier to find items as can be seen in the details below each. Better materials lead to better quality ferrofluids, but the procedure can be very flexible.

Reagents

• Hydrochloric acid (HCl) 5% - 31% concentration (Required volume will vary)
• This can be found as a cleaning agent at some hardware stores. It can be a little difficult to find and is restricted in some countries outside of the US. In theory, other acids can be substituted, but experiment at your own risk.
• Cleaning Ammonia (NH₄OH) 10%+ concentration
• Found as a cleaning solution. Needs to be unscented, and 10% ammonia can often be found as a "heavy-duty" variant but will be a little more expensive than lower concentrations.
• Ferric Chloride (FeCl₃) AND/OR Ferrous Chloride (FeCl₂.)
• Often found as a copper etchant for PCB making, this is the primary reagent as the source of the iron in ferrofluid. Anhydrous powder can be found here. Both ferric chloride (Iron (III) chloride) and ferrous chloride (Iron (II) chloride) are required for this procedure, but they can be made from one another. I have a separate Instructable on how to produce ferrous chloride from iron metal and hydrochloric acid here. I will expand on how either can be used when preparing the precursor solutions later.
• Oleic Acid
• This specific fatty acid works to form a surfactant on the nanoparticles, and is relatively cheap online. Though I haven't done extensive testing with it, olive oil contains about 80% oleic acid, which may work as a more easily available reactant here if pure oleic acid cannot be obtained.
• Distilled Water
• Ensures the product is as pure as possible as some of the reactions do involve salts and could react with minerals present in some tap water.
• Isopropyl Alcohol (91%+)
• Used for cleaning the nanoparticles towards the end of the preparation. Might be able to be filtered and reused.
• Steel wool
• A cheap, high surface area form of iron metal used in reducing ferric chloride (FeCl₃) to ferrous chloride (FeCl₂) if that route is taken in the procedure).
• Kerosene
• Used as the medium for the ferrofluid. Other related nonpolar solvents may work, but kerosene is tried, tested, and works wonderfully.

Equipment

• Magnetic or free-standing stirrer
• Fast mixing the during the magnetite formation step is essential to ensure small nanoparticle sizes. Magnetic stirrers work, but aren't ideal, because the stir bar can attract some of the magnetite which prevents it from being properly treated. A free-standing mixer with a chemical-resistant mixer would be better, but is more specialized and can be more expensive. I've even managed to get good quality ferrofluid by using this tabletop drill press with a custom-made mixer shaft.
• Borosilicate stir rod
• Useful for doing more manual mixing and decanting throughout the procedure. Especially useful when cleaning the magnetite. This could easily be replaced with a disposable plastic spoon or other small, non-porous mixing tool though.
• Pipettes – glass and/or disposable
• Useful for adding small amounts of a liquid which are needed in a few steps. Disposable pipettes are cheaper but do not last as long in the long-term.
• 50mL beaker
• 100mL beaker
• 250mL beaker
• 1000+mL beaker (small enough for stirrer to handle)
• For all beakers, they can be replaced by plastic containers of roughly equal volume. Beakers are preferred because they are less likely to stain, and have volume markers which may need to be added manually to a plastic alternative for some steps.
• 100mL borosilicate graduated cylinder
• Very useful for measuring liquids in some of the more precise steps. Avoid using an alternative material because both caustic and acidic liquids will be measured in it.
• Spray bottle
• Any general-purpose spray bottle should do. Will hold an ammonia solution, which is not particularly reactive with plastics. Using a spray bottle is critical to the quality of magnetite.
• Large neodymium magnet
• Useful for pulling the magnetite out of solution when cleaning it. The larger the magnet, the more ferrofluid you can process at one time.
• Glass crystallizing dish
• Used for the drying process of magnetite and the mixing of ferrofluid. Can be plastic, but kerosene damages plastics over time so they shouldn't be reused. Other materials like ceramic or metal may work, but they will likely stain.
• Vacuum chamber (Optional)
• Used for drying the magnetite to remove as much water as possible. The sun will work if a vacuum chamber isn't accessible.
• Appropriately rated glass or plastic containers to store magnetite nanoparticles and ferrofluid
• More notes later on about ferrofluid handling, but long story short, it's incredibly messy. Try to find some sort of glass container that easily stays upright and preferably has no rubber seal. I've had success with canning jars and test tubes.
• Waste fluid containers
• Waste fluids contain water with dissolved salt and soap, oils, bits of magnetite, and isopropyl alcohol. Everything except >5% alcohol should be fine to pour down the sink or disposed of outside, especially considering the low concentrations required for the procedure.
• Filter Paper
• Scale with precision 0.1g and limit >1kg
• Important for precise measurements of reactants, especially when proportions are important.
• pH paper
• Nitrile Gloves and safety goggles
• Only basic PPE is needed for this experiment. Hydrochloric acid, ammonia, ferric/ferrous chloride, and kerosene are all

A great explanation of how a true ferrofluid works can be found in a video on the channel "Fluids and Soft Matter UvA." Essentially, a ferrofluid consists of incredibly small magnetite particles, with diameters often 1/100,000th of a millimeter. These magnetite nanoparticles, which are ferromagnetic (attracted to a magnet but not magnetic itself), are then coated in a chemical which prevents them from grouping together or settling out of their carrier fluid. In this synthesis, the magnetite nanoparticles are coated in a fatty acid (oleic acid), which has a polar (hydrophilic) head and a nonpolar (hydrophobic) tail. The polar head is attracted to the magnetite particle while the nonpolar tail sticks out. With enough of these nonpolar tails, the entire particle acts nonpolar, and so osmotic pressure and Brownian motion within a nonpolar solvent like Kerosene are able to prevent these particles from interaction or falling out of solution, as if they were "dissolved." Thus, a stable colloid is formed and ferrofluid appears as a liquid, despite a large portion of its mass consisting of what would normally be a common type of rock.

How does this procedure make these specialized nanoparticles?

The procedure in this Instructable follows two main steps in forming these nanoparticles:

1. Forming magnetite nanoparticles via chemical reduction
2. Apply a fatty acid surfactant to form nonpolar nanoparticles

As seen in figure 1 of step 1, the solution in the reaction vessel consists of a ratio of two iron salts: ferric (Iron (III)) chloride (FeCl₃), and ferrous (Iron (II)) chloride (FeCl₂). These form a ratio of 2:1 because magnetite, Fe₃O₄, is made of two Fe³+ ions and one Fe²+ ion. This ensures that all the iron will be converted to magnetite, and not other non-ferromagnetic side products. To this (rapidly mixing) solution, ammonia (NH₄OH) is added through a spray bottle. This is to disperse the ammonia as much as possible and in combination with the rapid mixing, form as small of particles of magnetite as possible

As seen in figure 2 of step 1, the ammonia reacts with the iron chlorides, forming magnetite (Fe₃O₄) and a large amount of ammonium chloride (NH₄Cl) salt. This salt is essentially inert and is left in the solution throughout the duration of the procedure. Half of the hydroxide ions (OH-) that were originally bonded to the ammonium ions (NH₄+) split to become oxide ions (O²-) and free protons (positively-charged hydrogen (H+) ions). The protons bond with the other hydroxide ions to form water (H₂O), and the oxide ions form ionic bonds with the Iron ions to create the magnetite.

Figure 3 of step 1 represents the now-formed magnetite particles in the solution, which must be kept mixing until they're stabilized with surfactant in the second step.

As seen in figure 1 of step 2, an ammonium oleate soap, which is water-soluble because of its negatively-charged carboxyl group, has been added to the magnetite solution, with a single magnetite nanoparticle shown. After allowing the soap to mix with the magnetite nanoparticles, hydrochloric acid (HCl) is added to the solution. There is no strong interactions between the magnetite nanoparticle and the soap, because the soap has not yet reacted with the hydrochloric acid to become a fatty acid.

As seen in figure 2 of step 2, the hydrochloric acid has reacted with the ammonium oleate soap. This has formed more ammonium chloride, which remains aqueous, and replaces the ammonium ion bonded to the charged oleate with a hydrogen. By removing the charge on the carboxyl group of the oleate, the head of the fatty acid becomes less hydrophilic, and because the tail of the oleic acid is hydrophobic, it no longer becomes water soluble. Because the head is slightly more polar, it binds to the surface of the magnetite nanoparticles, causing the tails of the oleic acid to stick out. With enough of these tails, the nanoparticles are essentially hydrophobic, or nonpolar, themselves, and the particles are isolated from the rest of solution.

Figure 3 of step 2 represents the now-stabilized magnetite particles. These can then be isolated and combined with a nonpolar liquid which, unlike water, interacts with the tails sticking out of the nanoparticles, and allows the particles to be suspended.

Do this procedure if you only have Ferric Chloride available:

1.   In a 50mL beaker, add 0.02 moles FeCl₃ (3.24g anhydrous, or 5.40g hexahydrate, depending on the hydrate form), and add 20mL distilled water to make a 1M concentration solution.

2.   Weigh out an excess of steel wool (>0.01 moles or >0.55g). Add this to the 1M FeCl3 solution. Mix or let sit until the solution becomes a light green color and no more steel is consumed. It is recommended to stir to speed up the process. In this reaction, the FeCl3 is being reduced to a green-colored FeCl2. However, this reaction will slowly revert with exposure to oxygen in the air, forming various iron hydroxides and oxides in solution.

3.   Remove excess metal by filtration or careful decanting if needed and add more distilled water until there is 30mL of solution, now a 1M solution of FeCl₂. Place a low surface area iron source, like a nail, in solution until the rest of the precursor solutions are made to keep the iron at a 2+ oxidation state.

4.   This solution will now be denoted as “Solution A.”

Alternative Preparation if you have only Ferrous Chloride:

1.   Add 3.80g anhydrous FeCl₂ to 20mL of distilled water, stir until dissolved.

2.   Add more distilled water until the volume reaches 30mL.

3. Solution Should be a green-ish color. Place a low surface area iron source, like a nail, in solution until the rest of the precursor solutions are made to keep the iron at a 2+ oxidation state.

4.   This solution will now be denoted as “Solution A.”

1.   Add 5g of oleic acid to a 100mL beaker.

1.   Add 40mL of distilled water.

3.   Add 10mL of 10% ammonium hydroxide (by mass), or equivalent if maximum concentration is lower

4.   Stir until a thick, soapy, homogeneous mixture forms.

5.   This mixture will now be denoted as “Mixture B.”

1.   Prepare ~150mL of 5% hydrochloric acid (by mass) in a 200mL beaker from available concentration of hydrochloric acid and distilled water. More concentrated hydrochloric acid will need to be diluted down to the appropriate concentration. This volume is approximate because this solution is used to neutralize the reaction solution.

2.   This solution will now be denoted as “Solution C.”

1.   Add 9.73g of anhydrous FeCl₃ (or 16.21g hexahydrate) to ~450mL of distilled water in a 1000mL beaker - (the greater volume, the better. Larger containers and therefore greater amounts of water lead to lower solute concentration, potentially decreasing nanoparticle size and therefore the quality of the product. Feel free to use larger container if the stirring apparatus is able to support it).

Alternative Preparation

1. If only FeCl2 is available, add 7.61g of anhydrous FeCl2 (or 11.93g tetrahydrate) to 43.8g of 5% hydrochloric acid. Slowly add 34.1g of 3% hydrogen peroxide so that the FeCl2 will oxidize into FeCl3. Alternatively, instead of the hydrogen peroxide, air can be bubbled through the solution until the solution no longer changes colors. It should go from a green color to a dark brown or orange. This solution can then be added to to ~400mL of distilled water in a 1000mL beaker - (the greater volume, the better. Larger containers and therefore greater amounts of water lead to lower solute concentration, potentially decreasing nanoparticle size and therefore the quality of the product. Feel free to use larger container if the stirring apparatus is able to support it).

2.   Add 30mL of Solution A to make a 2:1 ratio of FeCl₃:FeCl₂ – (effective ratios may range from 2:1 to 1.7:1, so the ratio given may not be exact for the conditions of every synthesis. It varies based on the expected amount of additional oxidation of FeCl2 that occurs while mixing. Most oxidation can be avoided by completing the magnetite synthesis quickly).

4.   Prepare ~60g (42.6g stoichiometrically, [0.24M of NH₄OH] rounding up for excess) of a 10% solution of ammonium hydroxide (by mass) and dilute to about 100-200mL. Add it to a spray bottle and set nozzle to a fine mist.

5.   Cover the beaker if possible, and using a (preferably mechanical) stirrer, begin stirring the iron chloride solution as quickly as possible without causing significant surface splashing (attempting to prevent further oxidation of the Fe²⁺ ions into Fe³⁺ from dissolving oxygen gas).

6.   Using the spray bottle, quickly add the ammonium hydroxide, attempting to distribute the mist across as much surface area of the solution as possible. Color should transition from light yellow/orange to dark orange to brown, to black, becoming increasingly opaque. This step should be completed in under 2-5 minutes. Ensure pH of solution is above 9-10 and stop even if the bottle isn’t empty when the target pH is reached.

7.   Add the prepared Mixture B, ensuring it mixes well with the solution. Allow it to mix for 2-3 minutes or until the solution is visibly homogenous, whichever occurs later.

8.   Ensuring no visible changes are occurring, begin adding Solution C via spray bottle in the same way ammonium hydroxide was added. Fumes of ammonium chloride will be generated, but are harmless. As pH decreases below 7, the solution will visibly change. Black precipitate should begin coagulating and separating from solution, leaving a cloudy but colorless solution (of unbonded oleic acid and ammonium chloride). Continue adding hydrochloric acid until pH is around 5 to ensure all ammonium oleate has been destroyed, at which point turn off stirring.

9.   The nanoparticles are now stable, and synthesis is complete.

1.   Using a large neodymium magnet, gather all the stabilized magnetite nanoparticles to the base of the reaction beaker, and pour off any solution. Avoid pouring any magnetite if there is too much for the magnet to handle.

2.   Fill the beaker with ~200mL distilled water and mix magnetite to wash it and dilute the remaining solution.

3.   Repeat steps one and two 2-3 more times.

4.   Allow magnetite to mix with water in the beaker. Pour about a third into an appropriately sized beaker. Using the same magnet, hold it to the bottom. Pour off whatever doesn't stick in 5 seconds to another similar beaker. Repeat with that beaker but pour the rest into waste container. Wash both beakers and return the rest to the first smaller one.

5.   Repeat this process 2-3 more times, shortening the time allowed for the magnetite to stick. This is to ensure poor quality magnetite does not wind up in the final batch, if at the cost of some yield.

6.   Once clean, mix and pour the remaining magnetite into a beaker designated for the clean magnetite.

7.   Repeat steps 4-5 twice with the other two thirds of magnetite from the original beaker.

8.     Once all magnetite is in the designated beaker, allow it all to stick to the magnet, then pour off water and keep upside down over an absorbent towel for 5-10 minutes.

9.   Turn back over, and wash with a small amount 90%+ isopropanol. Solution should be slightly cloudy, as the isopropanol is removing some of the excess surfactant. Pour off into a separate waste container (if isopropanol cannot be disposed of in the same way as water, or if it is planned to be recycled). You will see that the magnetite appears to clump together less, and will form a smooth surface around the magnet

10. Repeat step 9 twice more.

11. Allow magnetite to dry completely. The ideal drying method is under vacuum but leaving it in the sun for a few hours is sufficient. Prevent exposure to high heat.

12. Magnetite should now be a slightly sticky or powdery material. It is safe for storage or use at this point, and shouldn't degrade over time.

1.   To a shallow glass or sacrificial plastic dish, add a 1:1 ratio (by mass) (just an estimation, likely not ideal) of magnetite and kerosene.

2.   Use a glass stir rod to attempt to "dissolve" the magnetite. Kerosene will become black even if it hasn’t absorbed the maximum amount of magnetite. This process usually takes a few minutes.

3.   A magnet may be used below the dish to mix further, and spiking/viscosity quality may be determined at that time. Once most magnetite has been dissolved, finished ferrofluid may be transferred to a glass container.

Keep in mind Kerosene's high flammability. Kerosene can also induce expansion and damage to most rubber-based materials, can chemically damage or weaken plastic, and can create fumes, so storage must be in an appropriate container (such as a jar with metal lid or a test tube with rubber stopper kept vertical). Ferrofluid is extremely messy and will stain most materials it contacts. It will stain plastic and glass black, brown, or orange, and will be absorbed by fibrous materials, leaving them black and moderately ferromagnetic. This ferrofluid shouldn't degrade over time, nor should it need to be re-mixed if it's left undisturbed for long periods.

Now that you've completed the synthesis, have some fun with ferrofluid! Pour it into a glass container and watch as magnets manipulate its mesmerizing spikes and curves. Cover a small magnet with ferrofluid for an even more captivating interaction. Take your experimentation further by observing how this incredible substance behaves on different iron objects in a magnetic field or create a desktop display by following some of the various tutorials on Instructables or around the internet.