Team:TU Delft-Leiden/Project/Gadget

Once the electrons are pushed into the counter electrode, the current generated is to be sensed. Since the myRIO /myDAQ require voltages to work with, the current is passed through a Current – to – Voltage convertor to obtain a voltage calibrated wrt the sensed current. Based on the range of the current to be sensed, the value of a certain precision Resistance is to be altered suitably. This has been discussed explicitly in the next section.

The circuitry discussed till now belongs completely to the domain of analog electronics. In order to store the sensed data onto a computer, the information has to be “digitized”. This is done with the help of an analog to digital converter (A to D). Similarly when random waveforms are produced on a PC, a digital version is initially created. Only when this digital data is passed through a digital to analog converter (D to A) does one obtain the actual waveform suitable for exciting the W.E. The common job of a D to A and A to D converter can be accomplished either by the DAQ or myRIO. Moreover, as these devices readily interface with a PC, real-time measurement data can be easily relayed to the PC screen.

A modular overview of the associated circuitry has been presented until now. The next section deals with the detailed analysis and functioning of the implemented circuitry. Before continuing, it is advised that the reader is familiar with the basic working of electrical /electronic circuits and especially OPAMPS. He /she must be familiar with basic principles namely Ohm's law, Kirchhoff’s laws, Norton – Thevnin equivalence etc. Before the next section begins, the same daunting schematic has been presented again but now with the different modules demarcated. It should feel easier to digest at this stage.

Detailed Circuit Analysis

Once the overall understanding of the system is achieved, the detailed working of the hardware can now be investigated. Operational Amplifiers (OPAMPs) form the most important component of the designed circuitry as it is the only active element. Ideal OPAMPS are amplifiers with an amplification factor A = infinity. Also, the input resistance of the amplifier is ideally infinite and is depicted in the figure below (fig 12) as Rin. Rout is finite and hence a finite current can flow through it .There are two input terminals of the OPAMP – the positive and negative terminals.

The OPAMP in principal is supposed to ideally amplify the difference in voltage between the two, which is called the differential input voltage. The terminals in figure 12 are as follows:

• V+ → Non inverting (Positive) input terminal
• V - → Inverting (negative) input terminal
• Vin → Net input Voltage into OPAMP = V+ - V-
• Vs+ / Vs- → Positive and Negative Power supplies
• Vout – Voltage output produced by OPAMP
• Rin /Rout → Input / Output Resistances
• A – Amplification factor of the OPAMP

Practical OPAMPs have finite but high gains and input resistances of the order of 10^6. The relationship between the Input and Output Voltages in an open loop setting is Vout = A( V+ – V-). Such massive gains are achieved by cascading various amplifiers with low gains in series with one another. These devices are capable of power amplification and termed active devices as they need an external power source to operate ( Vs+ and Vs-). The external power is required to bias the individual transistors that make up the OPAMP thereby preserving the law of conservation of energy.

Voltage Buffer/ Follower

A voltage buffer as mentioned before is used to regulate the output voltage around a certain fixed value, irrespective of the current drawn/ load attached => (Value of downstream resistance RL). Hence the most appropriate device for such a situation is the OPAMP in an unity negative feedback configuration. This is as illustrated below:

From the above figure 13 it is clear that the entire Output voltage Vo is ‘fed back’ into the ‘negative’ input terminal. Consequently, the term “Negative Feedback” is utilized to describe such an interconnection. In such a configuration, the Output Voltage Vo perfectly follows the input Voltage fed into the Vi + terminal irrespective of any downstream activity. Unity refers to the situation of 100% Output voltage being fed back.

The relationship for negative feedback is: Output Voltage Vo = A /(1 + A) x Vi . If A is very high (≥ 10^6), then Vo = Vi as A/(1+A) ≅ 1

Two voltage buffers have been used in the circuit at strategic locations where downstream activity may potentially alter the line voltage and are expected to improve the overall performance.

Voltage Summer & Polarity Inverter

This device is required to maintain the WE voltage at a specified voltage with respect to the RE voltage. This also consists of an OPAMP in a modified negative feedback configuration in order to generate the virtual ground and thereby converting all incoming voltages (V1, V2) into currents corresponding to the resistance used (R). The added current is now forced to pass through the resistor in the feedback loop to generate back the entire summed voltage V1 + V2. The current is pushed into the feedback resistance (Note: also R) due to the existence of infinite input resistance between the two input terminals. Hence the current would prefer to flow through a path providing lesser hindrance.

Vo = V1 + V2. Due the existence of Virtual Ground, the voltage here = 0. Hence, I1 = (V1 - 0)/R1 = V1/R. Similarly I2 = V2/R . Consequently, net current flowing through the feedback resistor = I1 + I2 . Hereafter the Voltage created at the output is: Vo = 0 – R(I1 + I2) = - (V1 + V2). Note that the inverted sum is produced at the output.

The above figure (fig 15) shows the configuration of a polarity inverter. The Voltage fed in Vin gets inverted to –Vin at the output. Vout exactly equals –Vin if Rf and Rin are equal in magnitude. With a different ratio of Rf:Rin , the amplification factor of the OPAMP can be made to assume values other than 1. The derivation is quite simple and analogous to the previous case. It is this inverter which is placed after the summer and before Buffer 2 for correct circuit operation. In the block diagrams this unit is represented as an integral part of the Voltage Adder.

Current to Voltage Converter

The incoming current from the Counter electrode Iin needs to be converted into a corresponding voltage which can be sensed by the D to A converter. This measured voltage is proportional to the current being sensed. By varying a precision resistance placed in the feedback loop R, the range of the current to be sensed can be adjusted. As evident from the figure (fig 16) this is also a case of negative feedback where the current pushed into the feedback loop gets converted into the inverted voltage.

Vout = Iin x Rf. Vout can be made to drive an auxiliary load Rl. This is not the case in our situation!

Practical Implementation / Software

Once the signals have been digitized, the acquired signals must be charted on some kind of graph in the signal (voltage / current) vs time format. It is intended that this graph will get updated in real time as increasing no of samples are acquired. In order to achieve this, the associated software - LabView needs to be programmed accordingly.

In general, any general purpose computation & Data Acquisition software like Matlab, Maple etc. can be interfaced with the DaQ provided the associated toolboxes are present. If required the software can be coded on C / C++ such that a “*.exe” is created, which is a standalone program and needs no third party software to run. As the myDaQ is a National Instruments product and so is LabView , the interfacing between the two has already been accomplished to a certain degree.

LabView provides an advantage over other conventional script based programming software like C etc. as it works on Graphical Code (G Code). One simply needs to have the logic ready in his/her mind, which can be easily implemented by dragging and dropping various functional blocks and interconnecting them via “wires”. These “wires” exhibits the signal flow within the digital domain in a software setting. This corresponding graphical code for this application has been represented in the figure below (fig 17):

The above wiring diagram is equivalent to the software code one writes using scripts but the advantage here is that the logic flow can be graphically seen and error can be easily trouble-shooted. The DAQ Assistants boxes are the functional blocks which enables interfacing with the analog domain. The “DAQ Assistant” block is responsible for converting the digitized version of the WE Voltage to an actual analog signal ( step / triangular) which is fit for excitation.

Likewise the block “DAQ Assistant2” plays the crucial role of receiving the analog domain signals (like the sensed CE current) and converting them into a digitized version for use on the PC. The code embedded within these blocks has been supplied by NI. The charts are used for plotting the information in real time as the data vectors grow with time. The LabView interface consists of two main windows. One is the Block Diagram where the G code is implemented. The other is known as the “front panel” and is used to simulate the front panel of an actual instrument (in this case the Potentiostat). This is the window where the charts actually plot the values and one can read them off directly! This is shown as follows:

As mentioned earlier, the front panel plays an essential role. It forms the interface between the user and the device which has been presented here. Two different kinds of excitation signals are possible – Either a step DC voltage or a triangular waveform for cyclic voltammetry. The value of the DC Step can be adjusted via the knob. Other waveform settings for the cyclic voltammetry mode like signal phase, amplitude and frequency can also be adjusted via suitable input options on the left.

The various charts show the different signals being applied and sensed. This image was captured when an ongoing test with a dummy cell was being performed. Crucial settings such as sampling rate and no of samples (Buffer size) should be carefully adjusted. The sampling rate must be chosen in accordance with Shannon’s Theorem. Or else the signal being sensed / generated would not be accurate due to probable aliasing issues.

The program can be downloaded : here

Testing & Validation

An AutoLab Eco Chemie Dummy Cell 2 was used to validate and further tune the designed potentiostat. A dummy cell is a device which can replicate the dynamics of electro-analytical cells and is made using passive electrical components like resistors and capacitors in certain configurations. With the dummy cell the expected current at the CE (Counter Electrode) can be pre-calculated based on the applied input voltage at the WE (Working Electrode).

With this data in hand, the WE terminal of the dummy cell was excited with a voltage of 0.5V .The RE (Reference Electrode) terminal of the dummy is at system ground (0 V) and hence whatever voltage is applied via the DAQ, the exact same amount gets physically applied at the WE. The dummy cell has various modes which replicate different scenarios and can be seen clearly from figure above (Fig 19).

For our validation purpose it was connected using mode (e) where a 10k-ohm resistance is between the WE and the RE/CE. Hence the current at the CE is expected to be 0.5V / 10000 ohm = 50 μA upon application of 1V at the WE. When the device is connected the measured signal must have a deterministic mean of around 50 μA with a stochastic spread on either side (possibly with a low covariance). The experiment was carried on for 100 seconds at the sampling rate of 2Hz. Hence 200 data points were collected .and the results were as follows (fig 20) :

These results confirm that the potentiostat is performing as anticipated. Henceforth this device can be used in conjunction with the bio reactor / MF device to characterize bio bricks or perform other electrochemical experiments. In the future, if current values to be probed are in the nano-amperes level, the ambient noise would be significant and hence can corrupt the data. To circumvent this problem a low pass filter can be used before feeding the data to the plotter. This would filter out the unwanted high frequency noise and therefore make the plots smoother. Another alternative is to performing averaging by choosing a Hanning Window of appropriate width. Although it performs smoothing, it comes at the cost of probable loss of system dynamics. Therefore the window width should not be too large and one must arrive at a reasonable trade off.

Specifications

• Sensitivity range - >20 micro amps
• Excitation modes – Step and Triangular
• Step Max Amplitude – +1 V
• Triangular Amplitude - +/- 1V
• Triangle Frequency Adjustable ( Analogous to Scan Rate) – (100Hz max)
• Sampling Rate/ buffer size Adjustable – (100 KHz max)
• Displays : Applied voltage at Working electrode - WE + RE
• Displays : Sensed CE current & calibrated Voltage
• Data can be saved directly onto Microsoft Excel in a vector-wise format

Timeline - Photos , Videos and Disasters

References

[1] An Inexpensive Field-Portable Programmable Potentiostat - Ashwini Vittal Gopinath and Dale Russell. The Chemical Educator ISSN: 1430-4171

[2] Image source - http://www.mep.net.au/teachingresearch/MEP_TEACHINGRESEARCH_04.html

[3] NPTEL - Analog Electronics - The Indian Institute of Technology Madras (IIT Madras , Chennai ,India)