Dear MEA users,
I recently had the opportunity to test a new make of multi electrode arrays from MultiChannel Systems (MCS) , and I was impressed with the recordings I obtained from them, so I'd like to share my experiences with you. Note that MCS gave me these arrays as a personal favor because I asked them at Neuroscience whether I could try the new arrays. MCS did not ask me to write this review.
The new model consists of a smaller glass substrate glued onto a ceramic carrier. This design made it possible to use considerably thinner glass than before, which is good news for all those who like to use short working distances lenses on inverted scopes. Another advantage of the new carrier is that the pads to which the pre-amplifier connects with its gold fingers, are much thicker and softer, which makes for more reliable connections. I didn't even need to clean the pads of with alcohol before use!
The electrode layout has been improved as well. The basic design is still an 8x8 grid with the corner pieces missing, but whereas before one more recording electrode was sacrificied to make space for a reference (bath ground) electrode, the new design features a reference electrode located in one corner of the array and accessed through a separate connector glued on the ceramic carrier (see pictures below). When inserting the MEA into the pre-amplifier, a separate short wire is used to connect this to a ground pin on the pre-amp. While this adds a few seconds of extra fiddling, it does create a much more reliable connection. Another added feature is a set of three 100 um diameter electrodes located in the other three corners of the array, which are accessed through gold pads on the ceramic carrier. I have not yet attempted to record from those.
The recording array itself consisted of 30 um diameter electrodes with 200 um inter-electrode spacing. RMS noise levels were 3.0 +/- 0.3 uV in the pre-amplifier's 10 Hz - 5 kHz pass band. I haven't measured the electrode impedance. When plated with dissociated rat cortical neurons at 1,000 cells/mm2 , most electrodes registered signals from many cells, with spike-amplitude- to-noise ratios ranging up to 50. Looking only at amplitude, an average of 1.6 cells could be isolated per electrode, and many more cells could be recorded from.
Stimulating through the MEA also worked well: out of 12 electrodes selected by their level of spontaneous activity, 11 could be used to elicit activity using biphasic voltage waveforms of less than 1V amplitude. (Current pulses of 10 uA typically charged up the electrode to 1V in about 50 us, counting both resistive and capacitive parts of electrode voltage.)
One of the stated goals of constructing using smaller glass plates is to get more chips out of a wafer, thus reducing the production cost per MEA. But even without any cost reduction, the new design is a great improvement. Be sure to check it out once it becomes generally available! And if you are planning to attend SIMEA 2003, please visit us to see some of the science we're using these arrays for  .
2. Quick snapshot through dissecting scope showing layout of electrodes. Close-up (archive) photos of electrodes are available at the MCS website.
 MultiChannel Systems are at www.multichannelsystems.com.
 D A Wagenaar and S M Potter: Real-time multi-channel stimulus artifact suppression by local curve fitting. J. Neurosci. Meth. 120:2 (2002) 113-120. [pdf]
 D A Wagenaar; R Madhavan; S M Potter: Stimulating news for MEA enthusiasts. Accepted for poster presentation at SIMEA2003, Denton, Tx, March 5-8 2003.
 R Madhavan; D A Wagenaar; C-H Chao; S M Potter: Control of bursting in dissociated cortical cultures on multi-electrode arrays. Accepted for poster presentation at SIMEA2003, Denton, Tx, March 5-8 2003.
 S M Potter; T B DeMarse: A new approach to neural cell culture for long-term studies. J. Neurosci. Meth. 110:1-2 (2001) 17-24.