Robotic measurements in liquids

G. Beni

Robotic measurements in liquids

Gerardo Beni

Journal of Electroanalytical Chemistry and Interfacial Electrochemistry

visibility

…

description

4 pages

link

1 file

Physico-chemical properties in liquids often depend critically on the geom

J. Electroana?. Chem., 140 (1982) 137-140 Ekevier Sequoia S.A., Lausanne- Printed in The Netherlands Prew ROBOTIC 137 note MEASUREMENTS IN LIQUIDS G. BENI BeNLaboratories.Holmdel,NJ07733 (U.S.A.) (Received 16th July 1982: in revised form 30th August 1982) Physico-chemical properties in liquids often depend critically on the geom- zyxwvutsrqponm etry of the experiment. Some of the most common examples are: heat flows, diffusion rates, current distributions, concentration gradients etc. In all these processes the geometry of the experiment determines the distribution of the physical properties within the liquid. Knowledge of this distribution is of primary importance in explaining and predicting the effects under investigation. For example, in electrocatalysis [l] , knowledge of the potential distribution in the electrochemical cell is often necessary to predict and/or control chemical reaction rates. In spite of their great importance, spatial distributions of physicochemical properties in liquids are not easily measured. In fact, manual adjustment and displacement of probe positions are usually cumbersome and irreproducible. An example is the extreme scarcity of measurements [Z] of current distributions at rotating disk and ring electrodes, even though theoretical calculations of these distributions are abundant [3]. Second, arrays or matrices composed of many probes are not very practical to fabricate and their size and shape makes it difficult to access the region to investigate. Finally, and most importantly, manual adjustment of the probes precludes monitoring of the time evolution of the spatial distribution investigated. In most cases this evolution is the most important aspect of the effect under study. In this note we propose the’ use of small-scale robots to investigate the spatial distribution of physicochemical properties in liquids. We show that this is now a practical possibility. Using a microprocessor controlled robot-arm we have mapped a 3dimensional set of 216 points on cubic lattices of lattice constant > 0.5 mm and < 10 mm, at rates of the order of -60 s per set. We discuss applications of this technique to electrochemical measurements. To illustrate the idea, let us consider a specific example, i.e. the phenomenon of electrode shape changes during electrochemical reactions. Shape changes play an important role in a large number of technologically important effects. Corrosion, electrode machining, electropolishing, electroforming, electrodeposition and electrolysis are important processes where time variation of electrode shape is a crucial part of the electrochemical process. During corrosion [ 41, for example, the shape of the electrode changes in time, often unpredictably. The change affects the current distribution around it, which in turn affects the reaction rate. Therefore, in order to predict and/or control corrosion rates it is nec0022-0728/82/0000-0000/$02.750 1982Elsevie~SequoiaS.A. 138 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA essary zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA to determine the change in current distribution during change of the electrode shape. So far this problem has been tackled only theoretically [El]. It is very difficult, however, to make realistic theoretical predictions. The numerical solutions are calculated for simple, idealized geometries, whereas typically the electrodes change shape in an irregular way (because of the formation of pits, steps, cracks etc.). It is this irregular shape which results in irregular current distributions and makes the entire process highly irreproducible and unpredictable. In most cases of corrosion the kinetics and the electrode size are such that probing the potential region surrounding the electrode (on a scale of 0.5 mm and time scale of tens of seconds) would provide a very useful measuring tool for monitoring the process. Fig_ 1. Small-scale microprocessor controlled robot-arm probes the pH on a set of 216 points (forming a cubic lattice with cell constant = 1.0 mm) in the vicinity of a corroding L-shaped iron electrode. This type of measurements can be carried out using a small robot-arm, as illustrated in Fig. 1. The robot-arm is a Minomover-5 [S] ; a small, computer-controlled manipulator which has recently become available commercially. The 5jointed mechanical arm provides an unusual combination of dexterity within a sphere of -40 cm. Lift capacity is 225 g, speed of motion can reach 15 cm/s and the resolution is -0.25 mm. The Minimover- reproducibility has proved satisfactory for our type of application, although the system has not been tested extensively. Reproducibility of motion is a crucial problem of robotics [ 7 1. On the other hand, for our ap- 139 zyxwvutsrqpo plication reproducibility is not as crucial as in manufacturing, where thousands of pattern-cycles may be required daily. In our case, re-calibration after -100 measurements is the only requirement. An accurate zero-position calibration technique is easily set up with excellent reproducibility. The robot-arm can be controlled by a variety of microcomputers. For simplicity we have interfaced the arm with an Apple II+ [S] , which is quite adequate for the applications proposed here. We have developed an algorithm for mapping a cubic lattice of 216 points with cell constants > 0.5 mm. The scanning rate depends on the cell constant and on the weight of the probe gripped by the robot hand. A typical scan may take place in -60 s. By varying the cell constant the liquid can be probed on the most sensitive scale for the experiment. The algorithm we developed includes 4 sequences of steps: 1. Calibration; 2. Pick up of electrode at fixed site; 3. Probing sequence (216 points); 4. Repositioning of the electrode in the initial position. The last three sequences can then be repeated by programming the robot arm to pick up a different probe, e.g. first using a thermocouple, then a pH electrode etc. Thus, a complex task of 3dimensional data acquisition can be handled by a single robot, without human supervision. Since the response time of most measuring electrodes is < 10 ms, the acquisition rate is generally limited by the robot-arm speed. Similarly, spatial resolution is limted by the robot-arm minimum step, since electrodes with -10 pm tips are available:Both speed and spatial resolution of the robot, however, are sufficient for many practical applications. It should be pointed out that the moving probe may perturb the system, acting as a stirrer. In certain applications it may be necessary either to reduce the data acquisition rate or to design a probe of suitable shape to minimize this perturbation. In many situations, however, this motion of the probe induces a negligible effect. For a specific example, in Fig. 1 the robot-hand grips a pH measuring electrode (resolution -0.1 mm) dipped in a solution containing a corroding sample of iron. Oxidation and reduction processes occurring during corrosion alter the local pH of the solution which is recorded at each of the 216 points probed by the robot. The distribution of pH is highly inhomogeneous due to the L shape of the iron sample. It is also time dependent due to the shape changes undergone by the sample during corrosion. Heating the solution further complicates the spatial distribution of acidity and temperature. Both these quantities can be recorded in succession using the technique described above. Results of these type of experiments will be reported elsewhere zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQP [S] . In this note we have shown the general method of robotic 3dimensional mapping of physico-chemical properties of liquids. Although so far robots have been used almost exclusively for manufacturing, we hope to have alerted the reader to the practical possibility of using small-scale robots for many measurements in physicochemical research. 140 ACKNOWLEDGEMENT zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA I thank S. Hackwood for much useful discussion and valuable comments. REFERENCES K.J. Vetter, Electrochemical Kinetics, Academic Press, New York, 1967. B. Miller and MJ. Bellavance, J. Electrochem. Sot., 120 (1973) 42. J.S. Newman, Electrochemical Systems, Frenti~Hall, Engiewood Cliffs, 1973. J.O’M. Bockris, B.E. Conway, E. Yeager and R.E. White, Comprehensive Treatise of Elec taochemistry, Vol. 4, PlenumPress,New York, 1981. R. AU&e, T. Begh and R.L. Sani, J. Electrochem. Sot., 125 (1978) Registered trademark of Microhot Inc. Proceeding of Robot VI Conference, Detroit, March 2-4,1982. Registered trademark of Apple Computers. S. Hackwood and G. Beni, to be published. 1981.

Log In

Robotic measurements in liquids

Related papers

Related papers