A primer to extracellular recording
Electric current flows in the tissue around the neurons during action potentials can be detected by means of extracellular microelectrodes as extracellular ‘spikes’. Extracellular spike potentials recorded from the mammalian central nervous system have a duration of between 0.2 and 10 ms. Their amplitudes are typically a few hundred microvolts although they may vary in amplitude from the noise level of the electrode (several microvolts) up to several millivolts, depending on the type of neuron and the quality of the recording system. The greatest advantage of extracellular recording is that the activity of neurons can be recorded without having to impale and consequently damage them. For this and other reasons, most in vivo neuronal spike detection is done with extracellular recording.
Signals picked up by extracellular electrodes are in the microvolt range and they need to be amplified to be able to be processed in more conventional electronic devices such as oscilloscopes, analyzers or computers. The usual degree of amplitude amplification in extracellular amplifiers is around 10,000x. The microelectrode must be connected to a specialized preamplifier (also known as a headstage probe) in order to work properly. The electrical properties of the headstage and main amplifiers set a limit to the smallness of signals that can be reliably measured.
The main difficulties with extracellular recording are the unwanted electrical signals or “noise” as it is popularly called. Noise in the present sense refers to unpredictable spontaneous voltage fluctuations, which appear as a thickening of the baseline when viewed on an oscilloscope at low sweep speed. An important distinction should be made, however, between the interferences (hum) from the mains power supply and the random noise that stems from the intrinsic properties of the substances from which the electrode and the electrical circuit are made. Alternating electric fields give rise to capacitive coupling between the power lines and the microelectrode input circuit, and this coupling is responsible for a considerable proportion of the interference problems because of the high impedance of the microelectrode. Interference from alternating magnetic fields is another pernicious problem that can obscure microelectrode recordings. Thirdly, ground (earth) loops occur when ground leads are connected in such a way that a loop or cyclic path for current exists. A ground loop can act as a tuned circuit to pick up line frequency magnetic field radiation causing line hum interference. Transformers in line-powered equipment, microscope lamps and their associated wiring commonly cause both magnetic and electric field interference. Ground loops can form when a number of amplifiers are powered from a common direct current (DC) supply; they can be several meters in circumference if they go through ground leads to line-powered equipment.
There are four main sources of random noise in the input circuit for microelectrode recording: (1) the Johnson or thermal noise of the resistance of the microelectrode, (2) the voltage noise of the headstage preamplifier, (3) current noise of the preamplifier, and (4) “excess noise” in the microelectrode. The Johnson noise is due to the thermal motion of electrons, and sets a lower limit to the total noise. The voltage noise inherent in the preamplifier is the noise measured at the output when the input is grounded. The current noise comes from the miniature current that all operational amplifiers (op amps) have to draw in order to measure the voltage generated by the electrode. This current has a steady component (DC bias or leakage current), onto which random fluctuations are superimposed. The noise current flows through the parallel combination of electrode resistance and stray capacitance, producing a noise voltage. Microelectrodes display a noise component that is additional to their Johnson noise. This excess noise depends strongly on the voltage applied to the microelectrode, even though some excess noise is present in the absence of any applied voltage. The current noise is linearly dependent on the resistance of the electrode in contrast with the square root dependence exhibited by the Johnson noise. Both types of noise are band-limited by the low-pass filter consisting of the resistance of the electrode and the total input capacitance.
Not much can be done about the random noise, except to use electrodes with lower impedances at the frequencies necessary for spike recording. Careful selection of high-quality op amps and other parts is also of crucial importance in amplifier designs. The key specifications for the op amps are the bias current and voltage noise. Adherence to a few design rules, however, can greatly diminish the interferences resulting from alternating electric or magnetic fields and ground loops. Appropriate grounding and shielding in combination with a prudent physical layout of the experimental set-up eliminates the need for the fully enclosed Faraday cage of yesteryear.
Microiontophoresis is most often used in conjunction with extracellular recording of neuronal firing. Spikes can be recorded through a multibarrel pipette assembly if one barrel is filled with a suitable electrolyte solution such as sodium chloride. However, electrolyte-filled glass micropipettes are electrically very ‘noisy’. The solid-conductor microelectrodes such as the tungsten or carbon fiber electrodes, in contrast, show significantly less noise in extracellular recordings. Carbon fibers are 5-8 micrometers in diameter and they provide excellent signal-to-noise ratio recordings.
Here, we offer an ultralow-noise headstage and main amplifier system for extracellular spike recording. This amplifier can be operated from batteries, totally independently from mains power. Its unique headstage probe design allows direct connection of the microelectrodes, further reducing destructive electromagnetic interferences and at the same time serves as an electrode holder.
If the measured noise voltages are squared and the mean square is computed, the fluctuations of both signs contribute positively. The square root of this mean (RMS) is the usual way of expressing the magnitude of noise voltages. The RMS of the Johnson noise is (4kTRdf)1/2, where k is the Boltzmann constant (1.38x10- 23 JK-1), T is the absolute temperature in Kelvin, R is the resistance of the microelectrode in ohm, and df is the noise bandwidth in Hz. If df is 5 KHz, this formula gives 5.6 microV RMS for thermal noise in the case of carbon fiber electrodes with a resistance of 0.4 MOhms when the temperature is 20 oC= 293 oK.When the input of our headstage amplifier is grounded, the voltage noise in the whole system is no more than 1 microV RMS. Using our four-barrel, carbon-fiber containing assemblies of borosilicate glass micropipets in the medulla of anesthetized rats, the total peak-to-peak noise levels is about 25 microV, i.e. about 6 microV RMS. This means that our headstage and main amplifier system is very close, in terms of noise performance, to its theoretical lower limit and does not contribute significantly to the overall noise.
The ExAmp-20KB, an affordable battery-operated extracellular amplifier
The ExAmp-20KB is a battery operated, AC-coupled amplifier designed for low-noise extracellular recording. Its unique headstage probe design puts the first stage of amplification (10x) at the microelectrode interface, resulting in less external interference noise pickup. The included 60 Hz (50 Hz optional) reject filter and our special electrode holder adapters virtually eliminate the need for Faraday cages. Possible applications are single-unit or single-axon recording in research or teaching experiments. It is not intended for use on human subjects. The ExAmp-20KB ships complete with headstage probe, operator‘s manual and batteries.
Specifications
Input impedance: 10 TeraOhm
Input leakage current: 0.8 pA
System gain with headstage probe: 2,000x to 20,000x
Filter bandpass frequencies: 300 to 8000 Hz
Output voltage swing: ±10 V, maximum
Power source: 4 standard D size batteries, included
Battery test: Audible tone
Lifetime of batteries: 500 hours (estimated)
Box dimensions: 6 1/8" x 2 1/4" x 6 7/8" (155 x 54 x175 mm) (WxHxD)
Weight: 2 1/4 lbs (1020 gram) (with batteries)
Order codes
For prices click on the catalog numbers.
M2100 ExAmp-20KB extracellular amplifier
Includes 1 headstage probe, 1 main unit, 4 “D” size batteries, 1 manual.
Optional:
M2110 Headstage probe for ExAmp-20KB.
M2331 Electrode holder adaptor for Kation-made CF microelectrodes
M2332 Electrode holder adaptor for metal microelectrodes
M2333 Electrode holder adaptor for 1.5 mm O.D. glass micropipette electrodes
The ExAmp-20K an affordable high-gain, low-noise extracellular amplifier
The ExAmp-20K is an AC-coupled spike amplifier designed for low-noise, monopolar extracellular recording. Its unique headstage probe design puts the first stage of amplification at the microelectrode interface, resulting in less external interference noise pickup. The built-in filters are optimized for single- unit recordings. The included 60 Hz (or 50 Hz for Europe) reject filter and our special electrode holder adapters virtually eliminate the need for Faraday cages. Front and rear panel controls are intentionally simplified for ease of use without compromising the quality of recordings. The ExAmp-20K can be used for research and teaching purposes only that do not involve human subjects. Requires an external power supply. Order separately.
| Specifications |
Front panel controls |
| Model and Catalog Number: | ExAmp-20KB, M2100 | Power: | Turns the unit on or off. LED signals on status |
| Input impedance: | 10 TeraOhm | Probe: | Receptacle for headstage probe |
| Input leakage current: | 0.8 pA | Gain: | Selects the magnitude of amplification |
| Probe gain: | 10x | Notch filter: | Turns on 60 Hz (or 50 Hz) reject filter |
| System gain: | 2,000x to 20,000x | Output: | Outputs for the amplified signal. |
| Bandpass frequencies: | 300 Hz to 8000 Hz | Rear panel controls | |
| Output voltage swing: | ±10 V, maximum | ||
| Power source: | External, 12 V DC power supply | Power In/Out: | 12V DC power connector |
| Power consumption: | 180 mA, maximum | Gnd: | Amplifier circuit`s floating ground |
| Amplifier (Gnd) ground: | Isolated from DC input ground (Case) | Case: | Housing and input power ground |
| DC power jack: | Rear panel, 2.1 mm x 5 mm type | ||
| Grounding receptacles: | Rear panel, miniature (2.64 mm) banana jack |
||
| Dimensions: | 155 x 54 x 175 mm | ||
| Weight: | 1.35 lb (614 grams) | ||
Order codes
For prices click on the catalog numbers.
M2200 ExAmp-20K extracellular amplifier
Includes 1 headstage probe, 1 main unit, 2 minature banana plugs, 1 manual.
M1116 External power supply with selectable US, EUR, GB or AUST plugs
Optional:
M2210 Replacement headstage probe for ExAmp-20K if needed.
M2331 Electrode holder adaptor for Kation-made CF microelectrodes
M2332 Electrode holder adaptor for metal microelectrodes
M2333 Electrode holder adaptor for 1.5 mm O.D. glass micropipette electrodes
The ExAmp-20KD, a two-channel extracellular spike amplifier
This modes is a dual-channel version of the ExAmp-20K extracellular amplifier with similar specifications. Requires an external power supply. Order separately.
Order codes
For prices click on the catalog numbers.
M2300 ExAmp-20KD extracellular amplifier
Includes 1 headstage probe, 1 main unit, 2 minature banana plugs, 1 manual.
M1116 External power supply with selectable US, EUR, GB or AUST plugs
Optional:
M2310 Replacement headstage probe for ExAmp-20K if needed.
M2331 Electrode holder adaptor for Kation-made CF microelectrodes
M2332 Electrode holder adaptor for metal microelectrodes
M2333 Electrode holder adaptor for 1.5 mm O.D. glass micropipette electrodes
Our headstage probe design is a classic, non-inverting amplifier with a 10x gain. In order to minimize interferences by alternating electromagnetic fields, microelectrodes can actually be plugged into the spark-plug type SMA receptacle of the headstage probe as shown below. Aminiature spring socket provides an input for the ground lead and mates with pin diameters ranging from 0.66 to 0.84 mm. The very short distance between the electrode and the first stage of amplification and the PTFE (Teflon) insulation of the receptacle allows a leakage-free, low-noise recording. The entire circuitry is placed in a 14 mm x 50 mm (diameter x length) nickel-plated cylindrical brass container for added shielding from electromagnetic interference. The center pin (inpu) is embedded in a PTFE insulator, whereas the grounding receptacle is in galvanic contact with the metal parts of the headstage probe. The headstage probe can be securely fixed in most micromanipulators via its 6.4 mm x 105 mm (diameter x length) mounting rod, which also serves as a cable guide. The headstage probe is connected to the main unit by a 1.3-m-long, well-shielded, high-flex cable through the use of a high-tech connector placed on the free end of the cable. The positive input of the front amplifier (center pin of the headstage’s input) is referenced to the ground (grounding socket) providing high-quality, single-ended (monopolar) recordings. This unique headstage probe design provides an electrode holder in one. See the adaptor section.
Physical layout of probe
Electrode holder adaptors
Three types of electrode holder adaptors can be screwed onto the input SMA connector. The first (No. 2331) is used to attach Kation-made, glass-insulated single or multibarrel carbon fiber electrodes. Microelectrodes can simply be inserted directly into the center pin of the SMA connector fixed in the front-end of the probe.
The second type of adaptor (No. 2332) is used to accommodate small-diameter metal electrodes. To configure this type of electrode holder, the connector pin of the electrode is inserted first into the front of the adaptor and pulled through it so that it protrudes somewhat from the back-end of the adaptor. This is done on a flat surface before the adaptor is screwed onto the SMA connector. A small pair of tweezers is used to grasp the pin of the electrode and insert it into the input socket of the probe. The adaptor is then screwed onto the probe gently by hand. Next, an about 2.5 cm-long strip of copper foil with conductive adhesive tape (3M, Part No. 1181) is rolled cylindrically so that one-third of the 1" (25 mm) width of the tape is around the front-end of the adapter and two-thirds of it is around the shank of the electrode itself. Finally, the front-end of the tape is flattened with the thumb and forefinger and the excessive parts of the foil are cut off. The copper foil has a twofold purpose: it holds the electrode in place and it provides an extended shielding for smoother recording. A common electrical tape can also be used.
The third type of adaptor (Cat. No. 2333) is a machined aluminum housing used to hold a glass micropipette in position and provides an additional shielding for a very low-noise extracellular recording.
To use this adaptor, first, solder a 30 to 40 mm long chlorided silver wire into the provided gold-plated pin. Alternatively, solder the wire into the pin and then chloride it using diluted HCl and a battery. Insert the pin in the center socket of the headstage probe (panel A). Carefully screw the body of the adaptor on the probe so that the silver wire protrudes from it (panel B). Pull the electrolyte-filled micropipette through the adaptor’s nut and put on the gasket ring as shown in panel C. Then, holding the micropipette, nut and gasket together, introduce the micropipette into the adaptor’s bore so that the silver wire goes smoothly into the micropipette (panel C). Finally, screw the nut onto the adaptor’s body by hand (panel D). Make sure that the micropipette is firmly positioned in the adaptor.
Order codes
For prices click on the catalog numbers.
M2110 Headstage probe for ExAmp-20KB.
M2210 Headstage probe for ExAmp-20K
M2310 Headstage probe for ExAmp-20KD.
M2331 Electrode holder adaptor for Kation-made CF microelectrodes
M2332 Electrode holder adaptor for metal microelectrodes
M2333 Electrode holder adaptor for 1.5 mm O.D. glass micropipette electrodes
Recording was taken from a brainstem
neuron using an ExAmp-20KB in combination with a Carbostar-3 carbon
fiber electrode. Cell firing was evoked by iontophoresed NMDA using
a Union-36 iontophoresis pump. The amplified signal was sampled and
digitized at 50 KHz frequency by a National Instruments PCI-1200 data
acquisition board. |
 |
Grounding and shielding in extracellular recording
In order to minimize external noise pickup and interference, experimental set-ups (‘electrophysiology rigs’)
have to be correctly shielded and grounded. Overall shielding is most often provided by a Faraday cage (a
grounded wire mesh box surrounding the baseplate, microscope, preparation and headstage probe). Faraday
cages, however, do not necessarily remove interference from magnetic fields, which usually cause the most
problem. To minimize magnetic field interference, place the the preparation to be recorded as close as possible
to a large carbon steel (not stainless steel!) baseplate which should be connected to the ground pin of the
headstage probe. The more massive the plate, the more helpful it will be at deflecting magnetic field. A steel
slab1/2" (12 mm) thick and at least 20" (500 mm) on each side greatly improves the quality of extracellular
recordings. Always keep the lead wire from the electrode to the headstage probe as short as possible, less then
2" (5 cm). This is accomplished by our unique headstage probe design where microelectrodes can actually be
plugged straight into the probe.
Keep line-powered equipments as far away from the site of recording as it is possible. Everything near the
preparation (manipulators, microscope, stereotaxic apparatus, microdrives, lamps, heaters, etc.) should be
grounded to a single point, that is the ground pin on the headstage probe and to nothing else. Use of a “star”
formation grounding will minimize ground loops. Isolation of extracellular amplifiers from power line ground
prevents ground loop formation. This is accomplished by employing battery power in our ExAmp-20KB
amplifier. After following these basic rules, the removal of interference noise is normally a process of trial and
error, involving experiments with slightly different patterns of grounding and shielding. Remember that our
ExAmp-20K amplifiers permit a very smooth, low-noise recording. Thus, if your extracellular recording
shows a frustrating level of noise keep trying to find and eliminate the source.
Keep line-powered equipments as far away from the site of recording as it
is possible. Everything near the preparation (manipulators, microscope, stereotaxic
apparatus, microdrives, lamps, heaters, etc.) should be grounded to a single
point, that is the ground pin on the headstage probe and to nothing else.
Use of a "star" formation grounding will minimize ground loops.
Isolation of extracellular amplifiers from power line ground prevents ground
loop formation. This is accomplished by employing battery power in our ExAmp-20KB
amplifier. After following these basic rules, the removal of interference
noise is normally a process of trial and error, involving experiments with
slightly different patterns of grounding and shielding. Remember that our
ExAmp-20K amplifiers permit a very smooth, low-noise recording. Thus, if your
extracellular recording shows a frustrating level of noise keep trying to
find and eliminate the source.
Agnesi, F., Tye, S.J., Bledsoe, J.M., Griessenauer, C.J., Kimble, C.J., Sieck, G.C., Bennet, K.E., Garris, P.A.,
Blaha, C.D. & Lee, K.H. (2009).Wireless Instantaneous Neurotransmitter Concentration System-based
amperometric detection of dopamine, adenosine, and glutamate for intraoperative neurochemical monitoring.
J Neurosurg, Vol. 111, No. 4, pp. 701-11.
Brakel, S., Babb, T., Mahnke, J. & Verzeano, M. (1971). A compact amplifier for extracellular recording.
Physiol Behav, Vol. 6, No. 6, pp. 731-3.
Budai, D. (1994) A computer-controlled system for post-stimulus time histogram and wind-up studies. J
Neurosci Methods 51:205-211.
Budai, D. & Molnar Z (2001) Novel carbon fiber microelectrodes for extracellular electrophysiology. Acta Biol
Szegediensis 45:65-73.
Conn, P.M., ed. (1991) Electrophysiology and microinjection. Methods in Neurosciences, Vol. 4., Academic
Press: San Diego.
Gerhardt, G.A. & M.R. Palmer (1987) Characterization of the Techniques of Pressure Ejection and
Microiontophoresis Using In-Vivo Electrochemistry. Journal of Neuroscience Methods, 22 (2): p. 147-160.
Gottschaldt, K.M., T.P. Hicks & C. Vahle-Hinz (1988) A combined recording and microiontophoresis technique
for input-output analysis of single neurons in the mammalian CNS. Journal of Neuroscience Methods,
23 (3): p. 233-240.
Hatzopoulos, A. & Theophilidis, G. (1984). A simple electronic unit allowing extracellular recording and stimulation
through the same wire hook or suction electrode. J Neurosci Methods, Vol. 11, No. 3, pp. 169-72.
Heredia-Lopez, F.J., Bata-Garcia, J.L., Gongora-Alfaro, J.L., Alvarez-Cervera, F.J. & Azpiroz-Leehan, J.
(2009). A digital programmable telemetric system for recording extracellular action potentials. Behav Res
Methods, Vol. 41, No. 2, pp. 352-8.
Imfeld, K., Neukom, S., Maccione, A., Bornat, Y., Martinoia, S., Farine, P.A., Koudelka-Hep, M. &
Berdondini, L. (2008). Large-scale, high-resolution data acquisition system for extracellular recording of
electrophysiological activity. IEEE Trans Biomed Eng, Vol. 55, No. 8, pp. 2064-73.
Jochum, T., Denison, T. &Wolf, P. (2009). Integrated circuit amplifiers for multi-electrode intracortical recording.
J Neural Eng, Vol. 6, No. 1, pp. 012001.
Kim, K.H. & Kim, S.J. (2000). Noise characteristic design of CMOS source follower and voltage amplifier for
active semiconductor micro-electrodes for neural signal recording. Med Biol Eng Comput, Vol. 38, No. 4,
pp. 469-72.
Kim, K.H. & Kim, S.J. (2000). Noise performance design of CMOS preamplifier for the active semiconductor
neural probe. IEEE Trans Biomed Eng, Vol. 47, No. 8, pp. 1097-105.
Land, B.R., Wyttenbach, R.A. & Johnson, B.R. (2001). Tools for physiology labs: an inexpensive highperformance
amplifier and electrode for extracellular recording. J Neurosci Methods, Vol. 106, No. 1, pp.
47-55.
McClellan, A.D. (1981). Extracellular amplifier with bootstrapped input stage results in high common-mode
rejection. Med Biol Eng Comput, Vol. 19, No. 5, pp. 657-8.
Nam, Y., Brown, E.A., Ross, J.D., Blum, R.A.,Wheeler, B.C. & DeWeerth, S.P. (2009). A retrofitted neural
recording system with a novel stimulation IC to monitor early neural responses from a stimulating electrode.
J Neurosci Methods, Vol. 178, No. 1, pp. 99-102.
Obeid, I., Nicolelis, M.A. &Wolf, P.D. (2004). A low power multichannel analog front end for portable neural
signal recordings. J Neurosci Methods, Vol. 133, No. 1-2, pp. 27-32.
Purves, R.D. (1981) Microelectrode methods for intracellular recording and iontophoresis. New York: Academic
Press.
Schrlau, M.G., Dun, N.J. & Bau, H.H. (2009). Cell electrophysiology with carbon nanopipettes. ACS Nano,
Vol. 3, No. 3, pp. 563-8.
Stamford, J.A., ed. (1992) Monitoring neuronal activity: a practical approach. 1992, Oxford University Press:
Oxford.
Uranga, A., Navarro, X. & Barniol, N. (2004). Integrated CMOS amplifier for ENG signal recording. IEEE
Trans Biomed Eng, Vol. 51, No. 12, pp. 2188-94.
Walker, T., Dillman, N. &Weiss, M.L. (1995). A constant current source for extracellular
microiontophoresis. J Neurosci Methods, Vol. 63, No. 1-2, pp. 127-36.
Wrobel, G., Zhang, Y., Krause, H.J., Wolters, N., Sommerhage, F., Offenhausser, A. & Ingebrandt, S. (2007).
Influence of the first amplifier stage in MEA systems on extracellular signal shapes. Biosens Bioelectron,
Vol. 22, No. 6, pp. 1092-6.
Yu, H., Cai, H., Zhang,W., Xiao, L., Liu, Q. & Wang, P. (2009). A novel design of multifunctional integrated
cell-based biosensors for simultaneously detecting cell acidification and extracellular potential. Biosens
Bioelectron, Vol. 24, No. 5, pp. 1462-8.
Zhang, X.,Wong, W.M., Zhang, Y., Gao, F., Nelson, R.D. & Larue, J.C. (2009). Design of a CMOS-based
multichannel integrated biosensor chip for bioelectronic interface with neurons. Conf Proc IEEE Eng
Med Biol Soc, Vol. 2009, No. pp. 3814-7.