RECIPE FOR METAL-PLATED RECORDING/INJECTION PIPETTES: This is an outline of procedures for making the metal-plated micropipettes described in:

GENERAL COMMENTS

I am not currently using these devices, and I have not bought new glass or insulating varnishes for many years, so any vendor information may be obsolete.

Some of the steps involved in cleaning the glass tubing and in surface preparation of the pipettes prior to plating are undoubtedly unnecessary. It is probably easier to continue this superstitious behavior than to identify the useless steps.

The most troublesome process is silvering the glass. If you have the facilities available, you might consider some other method of accomplishing this, such as vacuum deposition. Anything that provides a smoother coating than the chemical deposition methods would save a lot of grief in the subsequent steps. If you try some alternative method of silvering, I would appreciate hearing about your experiences.

It is convenient to have on hand 2 stock solutions before beginning the surface preparation (step 5). These are the silver reducing solution and the 10% KOH solution (step 6). Use distilled water for all procedures, including cleaning the glass. "Ethanol" is100% unless otherwise noted. All acids are in their fully concentrated form, unless otherwise noted.

Plan on spending a couple of long days on your first batch of 12 (exclusive of setting up equipment). After some practice, you might be able to cut this time in half, but only if the silver coating is smooth enough to allow adequate insulation with 2 or 3 coats.

I. GLASS

Use borosilicate (e.g., Pyrex) tubing whose outside diameter is twice the inside diameter. The insulation will add 30-50 microns to the outer diameter of the pipette, and this must be taken into account if the pipette is to fit closely into a guide tube. Close tolerances on the inside diameter are also important, since movement of the meniscus is used to determine how much has been injected. It is not terribly expensive to have glass tubing made to your specifications, but because the dimensions of any glass you get will vary from piece to piece, it is necessary to sort through the stock to pick out pieces that meet your tolerances. We use 0.031" o.d., 0.015 i.d. tubing, which fits smoothly into 18 ga hypodermic tubing after plating and insulating. My most recent supplier of custom glass is Friedrich & Dimmock, Inc., P.O. Box 230, Millville, NJ, 08332.

The completed pipette will have a metal-plated region whose length is determined by the structure being inactivated and the microdrive/micromanipulator system used (for most of the pipettes I've made, this is 90 mm). Above this will be a clear region whose length is determined by the maximum number of injections you will need to make in an experiment (typically, 80 mm for my experiments). The dimensions given here are those we have used for our experiments with acute, anesthetized preps, and these should be modified for your application.

PROCEDURES FOR MAKING THE RECORDING/INJECTING PIPETTES

1. Clean glass.

The purpose of this step is to clean the inside of each tube (the outside will be cleaned again in the surface preparation step prior to silvering). This is essential, since any junk in the tubing will be flushed down into the tip when the pipettes are loaded, leading to a high proportion of clogged pipettes. The cleaning is time-consuming, so clean enough glass for many batches.

Cut glass to 20 cm (or whatever length is appropriate for your experiment) and place in a graduated cylinder tall enough to allow the tubing to be completely submerged (for 20 cm glass, this is a 100 cc cylinder). Soak in hot detergent (I use a strong, filtered Alconox solution - a standard powdered lab detergent) for 20 min, and then flushed with large amounts of water introduced into the bottom of the graduated cylinder through a piece of glass tubing. This creates and upward flow that is efficient in flushing out the suds. Drain and fill with 95% ethanol. Heat in a double boiler until the ethanol boils (be careful!), and then immediately turn off the heat. After 15 min, drain, rinse well with water, and fill the graduated cylinder with fresh aqua regia (3 parts HCl to 1 part nitric acid).

Use a fume hood for the aqua regia, because it releases chlorine gas. As this happens, the solution will gradually turn from a pale yellow to a nasty-looking deep orange. Bubbles of chlorine gas will form within the tubes, and as they rise, they will cause aqua regia to be circulated through the bore of each tube. After the tubing has been in aqua regia for one hour, pour off most of the acid (wear gloves and use the hood!) into a large beaker half filled with water, and add water back to the graduated cylinder to cover the glass. This will stop the evolution of chlorine, making it safer to handle the diluted acid outside of the hood.

Flush liberally with water (again, directing the water to the bottom of the graduated cylinder though a glass tube), drain and dry in an oven at 150 C with the graduated cylinder on its side, capped with a small beaker (don't cap with aluminum foil - it will oxidize and coat your pipettes with powdered aluminum oxide) . When cool, store the cleaned glass upright in the graduated cylinder sealed with parafilm.

2. Pull the pipettes.

If you use a vertical puller, orient the glass so that the end that was down in the graduated cylinder is down in the puller, and discard the lower portion. If you use a horizontal puller, still discard the end that was down in the graduated cylinder. This is to prevent glass dust generated by contact between the tubing and the bottom of the graduated cylinder from being introduced into the pipette.

As each pipette is pulled, seal the rear end with a substance that will provide a water-tight barrier and withstand the solutions used in steps 13 and 14. I do this with powdered solder glass (the kind used for coating platinum-iridium microelectrodes) by poking the end in the powder, and then heating it in an alcohol flame until the powdered glass melts. If you do the same, take care not to tip the pipette in such a way as to allow the powder to fall into the pipette before it is fused. A better solution should be found for this, since the seal often breaks off.  Sealing by directly melting the end of the tubing does not work well because it tends to cause the tubing to bend, and because the end becomes rounded which makes it difficult to push into polyurethane foam for storage.

3. Heat-seal the tips.

Use an electrically-heated Nicrome or platinum wire. A straight segment of the heating element (1-mm diameter wire) in my old Kopf micropipette puller works well for this. Mount the wire horizontally and advance the pipette tip toward it from above with a stereotaxic carrier or a microdrive while viewing it through a stereomicroscope. Position the tip approximately 600-1000 microns above the wire and heat the wire to seal the tip. I control the heat with a variable transformer (roughly 60 volts AC), and the duration (start with 7 seconds) with a timer. Examine each tip under adequate magnification to be sure it is actually sealed.

4. Mount the pipettes for surface preparation and silver plating.

I use 25 x 150 mm test tubes to hold the solutions for these steps. Were I to set this up again from scratch, I would probably use a wider vessel to make it easier to introduce the pipettes without striking the side.  However, this would have the disadvantage of requiring larger volumes of all solutions. Whatever vessel you choose, it should be made of glass.

Twelve pipettes are held with a rubber band onto a grooved cylinder (mine is aluminum, nylon or delrin might be a better choice). Take care not to nick the rubber band when sliding the glass tubes under it (use a new rubber band for each batch!).

Mount the cylinder concentrically on a 1/4" metal rod that slides up and down in a Plexiglas bracket. The bracket locks onto the vertical post of a standard ring stand - I suggest clamping the base of the ring stand to the work surface. An adjustable stop on the upper end of the rod allows the pipette to be dipped to preset levels. A Plexiglas block with a hole to receive the test tube is rigidly attached to the base of the ring stand, positioned to centered the test tube under the cylinder.  When the cylinder is slid up, the pipettes are raised high enough to allow one test tube to be exchanged for another. Make provisions to lock the holder in the raised position without hassle. I tapped a short screw into the rod right above the holder, and then screwed a metal hook into the Plexiglas bracket positioned so that by rotating the rod, the screw engages the hook and holds the cylinder in its full-raised position. For insulating, the pipettes are slowly withdrawn from the test tube by a string passed across the hook and looped around this screw (see step 15).

The test tubes I use have the words "Pyrex" and "U.S.A." etched into them a few mm below the lip. I use these as a guide for filling the test tubes to standard levels for silvering and insulating. If your test tubes are not marked in some similar fashion, file a small horizontal line about 15 mm from the top. Mount the pipettes in the holder so that 92 to 95 mm will be submerged if the test tube is filled to the mark. This mark will define the level of the silvering solution only; other solutions used for surface preparation of the glass should be about 5 mm above this mark.

5. Clean the pipettes.

The main problem in developing these pipettes was getting adequate adherence of the silver to the glass. Surface preparation turned out to be the critical factor. The only procedures that I know are essential are the soak in hot KOH (which probably microetches the glass) and the stannous chloride sensitizing (step 7).

Soak in a strong Alconox solution for 10 minutes. Wash 3x in water. Soak 10 min in hot 95% ethanol. Wash 2x in water. Soak 10 minutes in nitric acid. Wash 2x in water. Soak 40 min in hot, concentrated KOH (100 g KOH made up to 125 cc with water). The KOH will get sufficiently hot when first mixed, but will not stay so for the entire 40 minutes, so keep a second tube with KOH heating in a double boiler and rotate the two every 10 minutes. After the full 40 minute soak, wash 3x in water. Check the tips at this point to be sure none are unsealed or broken. Any with broken tips will clearly have some fluid in the tip (magnification is not necessary). These should be discarded - they are useless, and once plated, they will be impossible to recognize.

6. Mix the silvering solutions.

I use the old Brashear's Process, as described on pg. 3428 of the 44th edition of the Chemical Rubber Company Handbook of Chemistry and Physics. This process was long ago abandoned for commercial silvering: it is expensive, and explosive compounds of silver are a side product (most mirrors are now aluminized anyway). On this scale, there is no danger - but unless you take care not to slop the silver solutions around, you will turn your hands, clothes, and much of your lab black.

Mix and keep on hand two stock solutions. For the reducing solution, dissolve 80 g sucrose in 700 cc water. Add 175 cc ethanol and 3 cc nitric acid. Add water to make the total volume 1 liter. Also, make up a stock bottle of 10% KOH in water.

The following solution should be made fresh for each silvering (I make it during step 5). Dissolve 1 g silver nitrate in 50 cc water. Add ammonium hydroxide (1 part to 3 parts water) drop by drop while stirring. A dark precipitate will form which will eventually begin to dissolve as more ammonium hydroxide is added. Keep adding until the precipitate is almost, but not quite, dissolved. Then add 5 cc of the stock 10% KOH. A dense precipitate will again form. Add ammonium hydroxide once more until the precipitate almost, but not quite, dissolves. Add water to make the total volume 150 cc. If a stir bar is used to make up this solution, cover the beaker with parafilm and introduce the ammonium hydroxide and KOH through a small hole to avoid splashing silver around.

7. Sensitize the glass.

This is essential for good adhesion of the silver. Make a fresh solution of stannous chloride by stirring a few grams in 100 cc water. Most will not dissolve, so filter the solution. Even after filtering, the solution will appear somewhat cloudy. Dip the pipettes for 4 or 5 sec, and then vigorously wash at least 3x in water to remove all particles.

8. Deposit a heavy coat of silver.

Position pipettes above an empty test tube. Mix 12 cc reducing solution with 48 cc silver nitrate solution and quickly pour into the test tube, filling it to the mark referred to in step 4. Immediately lower the pipettes into the solution and leave them undisturbed for 8 min. Wash 2x in water, and then with the pipettes raised above a beaker, rinse the remaining silver particles off by directing a stream of water from a wash bottle directly against each pipette. Wash again in water. If more silver particles float off, repeat the rinse with the wash bottle. When all loose silver has been removed, wash in ethanol and dry for 15 minutes. The test tube will also be silvered, and this can be removed with nitric acid (diluted slightly with water to prevent fuming). I always acid-clean all glass that has come in contact with the silvering solution.

This coating will be tough enough to allow the pipettes to be gently handled, but too rough near the tip for reliable insulation. Therefore, the tip is resilvered with a thinner coat in the next step.

9. Silver the tip of pipette with a thin coat.

Make up fresh silvering solution. Remove 3 mm of the silver from the tips by dipping them into a solution of 4 parts nitric acid to 1 part water. They must be dipped only for the few seconds required to remove the silver, or the acid fumes will damage the silver above the level of the acid (the reason the acid is diluted is to reduce fuming). Quickly rinse off the acid by washing 2x with water. Soak the lower 4 mm of the tips in hot KOH for 5 min. Wash 3x with water. Dip the lower 6 mm of the pipettes into the stannous chloride solution for 4 or 5 sec, then immediately wash 3x with water. Resilver the lower 6 mm for 110 sec, and immediately wash 3x with water. Use the wash bottle as in step 8 to remove loose particles of silver, wash once in ethanol, and dry for 15 minutes.

10. Examine the tips.

For the subsequent steps to work, the silver on the tips must be quite smooth. Examine a couple of the pipettes at an overall magnification of 200x. If the tapered portion of the pipette has large, jagged silver particles on it, the electrodes will be difficult to insulate. If this happens, one can remove the silver from the tip with nitric acid and repeat step 9. Alternatively, it is sometimes possible to gently polish the pipette tip with a cotton-tipped applicator soaked in ethanol. When is the silver too rough? Unfortunately, I can't quantify this - you'll only know this retrospectively, when you find that there are pin holes in the insulation that can't easily be corrected with additional coats of insulation. Note that the subsequent copper and gold plating will not correct excessive roughness in the underlying silver - the rough spots will be faithfully reproduced in the plated metal.

11. Remove excess silver.

This is to make all of the pipettes the same length, and to produce a clean edge on the silver. Fill a test tube to a depth of 90 mm with 4 parts nitric acid to 1 part water. Check to be sure that the solder-glass seal is still intact on each pipette. If it has been lost, it can be resealed at this stage. Dip each pipette into the test tube tail-end first to the bottom of the tube. The excess silver will be immediately removed. Rinse 2x in water.

12. Plate tip with copper

The silver is fragile, so the critical regions must be strengthened by electroplating. Most metals electroplated onto a conductive substrate will have internal stresses that would cause the plated metal to shrink slightly were it free to do so. Since the silver is very thin and does not adhere strongly to the glass, and if one electroplates gold or most other metals directly onto the silver, the silver will peeled off the glass. Fortunately, copper electroplates without such stresses, and will not strip the silver from the glass (this is why fine-quality silvered mirrors are copper-electroplated on the back - cheap ones are painted).

The plating solution is 70 g CuSO4.5H20 plus 2.5 cc sulfuric acid made up to 1 liter with water. Clean the lower 20 mm of the pipette with ethanol, followed by a water wash. Plate the lower 15 mm with the pipette negative relative to the solution. Do this one pipette at a time, using 5 volts in series with a 300 ohm resistor, and plating for 15 sec. Plate in a test tube lined with a spiral of a platinum wire connected to the positive terminal. A second short length of wire connected to the negative terminal is positioned over the mouth of the test tube, and as soon as the pipette tip is dipped into the solution, the silvered shaft of the pipette is touched to this piece to complete the circuit. I use a Kopf micromanipulator to hold and position the pipette for this step, although it can also be done freehand. Wash with water immediately after plating.

13. Plate copper on the point of electrical contact.

A segment of metal abutting the clear part of the glass tubing is left uninsulated to provide a point of electrical contact with the exposed ring at the pipette tip. This must also be plated to reinforce the silver. Plate this segment using the same strategy as in step 12, but with the pipette inverted. For my purposes, this segment is 15 mm long, and it is plated for 15 seconds.

14. Gold-plate the copper.

It is difficult to make a gold plating solution that will work satisfactorily, so it is probably best to use a proprietary solution. Satisfactory plating solutions can be obtained from jewelers. Clean the tip with ethanol, and wash with water. Plate the lower 10 mm of the pipette (to beyond the point where it reaches the full width of the shaft) using the same setup as for copper plating. I use 5 volts, in series with a 5000 ohm resistor, for 25 sec. Plate the rear 12 mm of the electrode with gold using the same procedure.

15. Insulate the pipette.

The goal is to insulate all of the plated metal except for the rear 10 mm, which is used for connections to the amplifier. This rear portion must be protected from oxidation during the subsequent baking steps. One would think that the gold would provide adequate protection, but does not. For a totally unrelated purpose (gold-plating copper rings for the double-induction method of measuring eye position), I recently found that the kind of gold plating processes typically used by jewelers cannot protect the copper from a soak in nitric acid. It always leaves numerous, invisible pin holes that allow rapid attack by the acid, and these may also account for the lack of protection from oxidation for the pipettes. In any case, the gold is protected by covering it with a substance that will withstand the heat but that can be removed afterwards. For this purpose, I use Insl-x brand insulation (Insl-x Products Corporation, Yonkers, N.Y.). This does not polymerize and it can be removed with solvent afterwards. I use Isonel brand synthetic varnish for the permanent insulation on the rest of the metal-plated surface (Schenectady Chemicals, Inc., Schenectady, New York.). This polymerizes when heated, and cannot be easily removed with solvents. The following procedure is designed to prevent any of the Insl-x from being trapped under the Isonel.

a. Arrange some means of lifting the pipette holder at a steady rate of about 10 mm/min. We simply draw it up with a string attached to a Harvard variable speed infusion-withdrawal pump (review step 5). Mount all 12 pipettes, wash them with xylene, and then lower them (while still wet) into a test tube filled with Isonel until only 10 mm of the gold-plated region at the rear end is exposed. Withdraw the pipettes slowly until the tips are all clear of the Isonel. If you withdraw too fast, the Isonel will bead. Once this happens, the beading will become worse with each subsequent coat of Isonel. Dry for 15 minutes.

b. With a small artist's brush, paint the rear 11 mm of the pipettes with Insl-x (i.e., slightly overlapping the Isonel), being certain that no metal is left exposed. Bake for 30 minutes at 150 C to polymerize the Isonel.

c. Clean the pipettes by dipping in xylene, taking care not to allow the xylene to come in contact with the Insl-x (Insl-x dissolves in xylene, but the Isonel is insoluble in organic solvents after it is polymerized). Dip the pipettes in Isonel to within 1 or 2 mm of the Insl-x, and withdraw them again at a controlled rate to put on a second coat of Isonel. Bake again at 150 C for 30 minutes.

d. Repeat step c at least one more time. As many as 6 coats may be needed if the silver is very rough, but more than 3 are seldom required.

PROCEDURES FOR USING THE PIPETTES:

1. Prepare the amplifier contact surface.

Remove the Insl-x from the amplifier contact surface with cotton-tipped applicators soaked in Insl-x thinner or xylene. This surface is too fragile to allow you to connect the amplifier with an alligator clip or some similar device. You need to arrange something that gently spring loads a smooth copper (or other conductive) surface against the contact point. You can strengthen the exposed plated metal on the pipette with conductive silver paint of the type made specifically for electronic circuits. However, although this provides a nominally stronger contact surface, if it fails and flakes off, it will strip away the underlying plated metal as well. I don't drive the pipette into the brain from this point of contact - it is too fragile. Contact for advancing the pipette is provided by clamping the hypodermic tubing on the upper-most part of the pipette (step 3) to an extension of the microdrive/micromanipulator. An alternative way to protect the contact surface might be to glue a 10-mm length of hypodermic tubing around it with either the silver paint or a conductive epoxy, but I've never tried this.

2. Break off the sealed part of the tip.

Break off some of the tip so that the outside diameter is about 15 microns. I've found that doing this by running the tip into some hard object, even under microscopic control, often results in the pipette breaking too far back from the tip. The first time I used these pipettes in a real experiment, I ruined 6 or 8 before getting one broken off to the correct diameter. Given the work involved in making them, if you use them with any regularity, you should devise a more reliable way of breaking them (or, if you have a very steady hand and sharp vision, try using a pair of iris scissors). I'll describe below the method that I use.

I break off the tips by pinching them between a short segment of a #10 scalpel blade (cut to the shape of an "axe head") and the edge of a rectangular piece of a cover glass. The blade and cover glass are held by manipulators attached directly to a microscope frame, whereas the pipette is moved relative to them with the microscope stage. The tip is viewed with a 10x objective, which provides a good balance between magnification and working distance.

Exactly how you do this depends on the configuration of your microscope and the equipment you have available. In my case, I devote an inexpensive microscope to the procedure. I hold the pipette on a fixture mounted on the rear slide of the original mechanical stage. The height of the pipette can be adjusted by turning a screw on the fixture, affording three-dimensional translation of the tip relative to the objective lens. To the right of the mechanical stage on most microscopes is a small vernier scale on a pedestal that is fixed relative to the microscope base. I removed the vernier from the pedestal and mounted onto the pedestal a second microscope mechanical stage (a cheap one bought from the Edmund Scientific Co.) which is used to independently move the "axe head". The rectangular piece of cover slip is held in position by a second fixture mounted on the same vernier pedestal so that one of the original (uncut) edges faces opposite the "axe head", serving as a "chopping block". The position of the "chopping block" is only crudely adjustable: it is configured so that it always appears in the field of view without need for adjustment. The scalpel blade and cover slip are oriented in the same plane (the axis of symmetry of the microscope), with the edge of the "axe head" perpendicular to the microscope stage. Since the "axe head" has some curvature, simple provisions are made to adjust its orientation so that it will contact the "anvil" roughly in the center of the cutting edge (glue the "axe head" to the end of a short rod that can be rotated via a lever attached to the rod).

In use, the "axe head" is advanced until it just touches the "anvil", and the microscope is adjusted to bring this point into clear view via the focusing knob. Then the "axe head" is withdrawn with the secondary mechanical stage, and the pipette advanced into the gap between the "axe head" and "anvil" with the primary mechanical stage. The height of the pipette is adjusted to bring it into good focus, which also puts it at the contact point between the "axe head" and "anvil". Then the "axe head" is brought back into contact with the "anvil" to nip off the tip at the desired location.

3. Prepare the rear of the pipette.

Break a couple of mm off the rear end of the pipette to remove the seal, and glue on an adapter made of hypodermic tubing. The hypodermic tubing will provide an air-tight seal to the P.E. tubing. This allows you to inject fluid from the tip by applying air pressure. The seal between metal and P.E. tubing is much better than obtained by fitting P.E. tubing directly to the glass, and the hypodermic tubing provides a gripping point for advancing the pipette via by a fixture attached to to microdrive/micromanipulator (step 1). This will also prevent the glass pipette from being broken should the P.E. tubing be accidentally tugged.

4. Attach a scale.

Glue a paper mm scale cut from some graph paper onto the pipette for measuring the amount injected. For the tubing that we use, a 1-mm movement of the meniscus is equivalent to about 115 nanoliters.

5. Fill the pipette.

Fill the pipette by pressure from the rear with a syringe, passing the fluid though a 0.2 micron filter. If the solution is to be refrigerated, warm it to room temperature first, or bubbles will form in the pipette. One difficulty is that you can't see what is going on in the metal-plated part of the pipette. However, you can check for bubbles there by applying pressure and then quickly releasing it. If the meniscus jumps back up when pressure is released, there is a bubble in the pipette.

6. Platinize the tip.

Just before use, platinize the tip by passing 400 nanoamps for 15 to 20 seconds, with the tip negative. When you reuse a pipette, it must be platinized again before every use.

Platinizing Solution:

0.5 g chloroplatinic acid

5 mg lead acetate

make up to 50 ml with dH20

add 0.5 ml 0.5% ungelled gelatin solution

7. Making injections.

Use a 5 or 10 cc plastic syringe to provide air pressure. Drill a small hole near the needle end of the syringe barrel. Cover the hole with a finger when making an injection - removing your finger releases the pressure immediately.

Occasionally, the pipette will clog. When this happens, using higher pressure (such as that provided by a gas tank) almost always fails to clear it. If you are using a guide tube, try withdrawing the pipette into the guide tube and injecting some fluid. This occasionally works, but usually it is necessary to withdraw the pipette from the brain to unclog it, or to replace it with another one. Try cleaning the tip as in step 7, and if this does not work, nip more glass off the tip. I have successfully used tips with an outside diameter of 50 microns, but larger tips record poorly and often leak excessively. For experiments with acute preparations, we configure our micromanipulator so that the guide tube can be left in the brain while a pipette is withdrawn or inserted.

My experience with these pipettes is entirely for deep brain structures, and for these, leaking is usually not a problem. However, on occasion the fluid leaks out spontaneously and rapidly, even with fairly small tips. I don't know the reason for this - it is as if there is a vacuum in the brain sucking out the fluid. The obvious solution is to put some suction on the fluid column, but whereas this stops the leak, it often causes the tip to clog.

8. Clean the tip.

When the experiment is over, clean the outside of the pipette with a cotton applicator dipped in a hot solution of lauryl sulfate (dodecyl sodium sulfate - SDS). It is necessary to mechanically clean the tip with the cotton to get all of the brain junk off. Then with the pipette tip submerged in the SDS, force the remaining solution out with air pressure. When a stream of bubbles appears, raise the tip out of the SDS and rinse it off with water while keeping air pressure in the pipette. Remove the pipette from the water and release the pressure.

When storing pipettes that have been opened and used, do not support them by sticking the rear end in foam rubber (or any other soft material), because fragments of foam will get into the pipette and subsequently clog the tip. With care, an individual pipette can be used for several preparations. When I was using these regularly, a batch of 12 would last 1-2 years.

Last change: February 25, 2000 Most recent glass supplier added.