Experiment 32 Galvanic Cells, the Nernst Equation -. Copper metal spontaneously oxidizes to copper(lll ion in a solution containing silver ion. Silver metal crystals form on the surface of the copper metal. -5 =< ----- • To measure the relative reduction potentials for a number of redox couples • To develop an understanding of the movement of electrons, anions, and cations in a galvanic cell • To study factors affecting cell potentials • To estimate the concentration of ions in solution using the Nernst equation OBJ E c TI v Es The following techniques are used in the Experimental Procedure: TE c HNI Q u Es Electrochemical cells are of two types, galvanic and electrolytic, both employing the principle of oxidation-reduction (redox) reactions. In galvanic (or voltaic) cells (this experiment), redox reactions occur spontaneously as is common with all portable batteries of which we are very familiar. Electric cars, flashlights, watches, and power tools operate because of a specific spontaneous redox reaction. Electrolytic cells (Experiment 33) are driven by nonspontaneous redox reactions, reactions that require energy to occur. The recharging of batteries, electroplating and refining of metals, and generation of various gases all require the use of energy to cause the redox reaction to proceed. Experimentally, when copper wire is placed into a silver ion solution (see opening photo), copper atoms spontaneously donate electrons (copper atoms are oxidized) to the silver ions (which are reduced). Silver ions migrate to the copper atoms to pick up electrons and form silver atoms at the copper metal-solution interface; the copper ions that form then move into the solution away from the interface. The overall reaction that occurs at the interface is: INTRODUCTION Cu(s) + 2 Ag+(aq)---> 2 Ag(s) + Cu 2+(aq) (32.1) This redox reaction can be divided into an oxidation and a reduction half-reaction. Each half-reaction, called a redox couple, consists of the reduced state and the oxidized state of the substance: + 2 e- oxidation half-reaction (redox couple) (32.2) 2 Ag+(aq) + 2 e- ---> 2 Ag(s) reduction half-reaction (redox couple) (32.3) Cu(s)---> Cu 2+(aq) Interface: the boundary between two phases; in this case, the boundary that separates the solid metal from the aqueous solution Redox couple: an oxidized and reduced form of an ion/substance appearing in a reduction or oxidation half-reaction, generally associated with galvanic cells Experiment 32 357 Potentiometer Cu Cu half-cell Cu(s)- Cu 2 •(aq) + 2 e- Ag Ag half-cell Ag•(aq) + e- -Ag(s) Figure 32, 1 Schematic diagram of a galvanic cell Half-cell: a part of the galvanic cell that hosts a redox couple External circuit: the movement of charge as electrons through a wire connecting the two half-cells, forming one-half of the electrical circuit in a galvanic cell Salt bridge: paper moistened with a salt solution, or an inverted tube containing a salt solution, that bridges two half-cells to complete the solution part of an electrical circuit Internal circuit: the movement of charge as ions through solution from one half.cell to the other, forming onehalf of the electrical circuit in a galvanic cell Cell Potentials 358 A galvanic cell is designed to take advantage of this spontaneous transfer of electrons. Instead of electrons being transferred at the interface of the copper metal and the silver ions in solution, a galvanic cell separates the copper metal from the silver ions to force the electrons to pass externally through a wire, an external circuit, for the reduction of the silver ions. Figure 32.1 is a schematic diagram of a galvanic cell setup for these two redox couples. The two redox couples are placed in separate compartments called half-cells. Each half-cell consists of an electrode, usually the metal (reduced state) of the redox couple, and a solution containing the corresponding cation (oxidized state) of the redox couple. The electrodes of the half-cells are connected by a wire through which the electrons flow, providing current for the external circuit. A salt bridge that connects the two half-cells completes the construction of the galvanic cell (and the circuit). The salt bridge permits limited movement of ions from one half-cell to the other, the internal circuit, so that when the cell operates, electrical neutrality is maintained in each half-cell. For example, when copper metal is oxidized to copper(II) ions in the Cu 2+1cu half-cell, either No3- anions must enter or copper(II) ions must leave the half-cell to maintain neutrality. Similarly, when silver ions are reduced to form silver metal in its half-cell, either NO 3- anions must leave or cations must enter its half-cell to maintain neutrality. The silver electrode at which reduction occurs is called the cathode; the copper electrode at which oxidation occurs is called the anode. Because an oxidation process donates electrons to the copper electrode to provide a current in the external circuit, the anode is designated the negative electrode in a galvanic cell. A reduction process accepts electrons from the circuit and supplies them to the silver ions in solution; the silver cathode is the positive electrode. This sign designation allows us to distinguish the anode from the cathode in a galvanic cell. Different metals, such as copper and silver, have different tendencies to oxidize; similarly, their ions have different tendencies to undergo reduction. The cell potential of a galvanic cell is due to the difference in tendencies of the two metals to oxidize (donate electrons) or of their ions to reduce (accept electrons). Commonly, a measured reduction potential, the tendency for an ion (or molecule) to accept electrons, is the value used to identify the relative ease of reduction for a half-reaction. A potentiometer or multimeter, placed in the external circuit between the two electrodes, measures the cell potential, Ece11, a value that represents the difference between the tendencies of the metal ions in their respective half-cells to undergo reduction (i.e., the difference between the reduction potentials of the two redox couples). Galvanic Cells, the Nernst Equation For the copper and silver redox couples, we can represent their reduction potentials as Ecu" ,cu and EAg\ Ag• respectively. The cell potential being the difference of the two reduction potentials is therefore (32.4) Experimentally, silver ion has a greater tendency than copper ion does to be in the reduced (metallic) state; therefore, Ag+ has a greater (more positive) reduction potential. Since the cell potential, E,elh is measured as a positive value, EAg\Ag is placed before Ecu" ,cu in equation 32.4. The measured cell potential corresponds to the standard cell potential when the concentrations of all ions are I mol/L and the temperature of the solutions is 25°C. The standard reduction potential for the Ag+(! M)/Ag redox couple, E° Ag•,Ag• is +0.80 V, and the standard reduction potential for the Cu2+(1 M)/Cu redox couple, E°cu" ,Cu• is +0.34 V. Theoretically, a potentiometer (or multimeter) would show the difference between these two potentials, or, at standard conditions, E°cen = E° Ag',Ag- E°cu" .Cu = +0.80 V - (+0.34 V) = +0.46V Silver ;ewelry is longer lasting than copper ;ewelry; therefore silver has a higher tendency to be in the reduced state, a higher reduction potential (32.5) Deviation from the theoretical value may be the result of surface activity at the electrodes or activity of the ions in solution. In Part A of this experiment, several cells are "built" from a selection of redox couples and data are collected. From an analysis of the data, the relative reduction potentials for the redox couples are determined and placed in an order of decreasing reduction potentials. In Part B, the formations of the complex [Cu(NH 3) 4]2+ and the precipitate CuS are used to change the concentration of Cu 2+(aq) in the Cu 2+1cu redox couple. The observed changes in the cell potentials are interpreted. Measure Cell Potentials The Nemst equation is applicable to redox systems that are not at standard conditions, most often when the concentrations of the ions in solution are not I mol/L. At 25°C, the measured cell potential, £,en, is related to E°cen and ionic concentrations by Measure Nonstandard Cell Potentials . E cell Nernst equat10n: = c,=cell - ~~nl og Q --n- (32.6) where n represents the moles of electrons exchanged according to the cell reaction. For the copper-silver cell, n = 2; two electrons are lost per copper atom and two electrons are gained per two silver ions (see equations 32. 1-32.3). For dilute ionic concentrations, the reaction quotient, Q, equals the mass action expression for the cell reaction. For the copper-silver cell (see equation 32.1 ): [Cu 2+] Q= [Ag+]2 In Part C of this experiment, we study in depth the effect that changes in concentration of an ion have on the potential of the cell. The cell potentials for a number of zinc---{:opper redox couples are measured in which the copper ion concentrations are varied but the zinc ion concentration is maintained constant. Zn(s) + Cu 2+(aq)-----, Cu(s) + Zn 2+(aq) Mass action expression: the product of the molar concentrations of the products divided by the product of the molar concentrations of the reactants, each concentration raised to the power of its coefficient in the balanced cell equation The concentrations (i.e., density) of solids are constant and therefore do not appear in mass action expressions. The Nemst equation for this reaction is E _ E° cell - cell - 2 0.0592 lo [Zn +] 2 g [Cu 2+] (32.7) Rearrangement of this equation (where E°cen and [Zn 2+] are constants in the experiment) yields an equation for a straight line: E cell -_ £ 0cell y= - 0.0592 - Iog [Z n2+ ] + -- 2 b + 0.0592 2 log [Cu2+] m (32.8) x Experiment 32 359 To simplify, pCu = -log {Ctl•j Ece 11 = constant - 0.0592 - -- pCu 2 (32.9) A plot of Ece 11 versus pCu for solutions of known copper ion concentrations has a negative slope of 0.0592/2 and an intercept b that includes not only the constants in equation 32.8 but also the inherent characteristics of the cell and potentiometer (Figure 32.2). The E,. 11 of a solution with an unknown copper ion concentration is then measured; from the linear plot, its concentration is determined. Procedure Overview: The cell potentials for a number of galvanic cells are measured and the redox couples are placed in order of decreasing reduction potentials. The effects of changes in ion concentrations on cell potentials are observed and analyzed. Perform the experiment with a partner. At each circled superscript@ in the procedure, stop and record your observation on the Report Sheet. Discuss your observations with your lab partner and your instructor. EXPERIMENTAL PROCEDURE A. Reduction Potentials of Several Redox Couples The apparatus for the voltaic cell described in the Experimental Procedure may be different in your laboratory. Consult with your instructor. 1. Collect the electrodes, solutions, and equipment. Obtain four small (-50 mL) beakers and fill them three-fourths full of the 0.1 M solutions as shown in Figure 32.3. Share these solutions with other chemists/groups of chemists in the laboratory. Polish strips of copper, zinc, magnesium, and iron metal with steel wool or sandpaper, rinse briefly with dilute (--0.1 M) HNO 3 (Caution!), and rinse with deionized water. These polished metals, used as electrodes, should be bent to extend over the lip of their respective beakers. Check out a multimeter (Figure 32.4) (or a voltmeter) with two electrical wires (preferably a red and black wire) attached to alligator clips. 2. Set up the copper-zinc cell. Place a Cu strip (electrode) in the CuSO4 solution and a Zn strip (electrode) in the Zn(NO 3) 2 solution. Roll and flatten a piece of filter paper; wet the filter paper with a 0.1 M KNO 3 solution. Fold and insert the ends of the filter paper into the solutions in the two beakers; this is the salt bridge shown in Figures 32.1 and 32.3. Set the multimeter to the 2000-mV range or as appropriate. Connect one electrode to the negative terminal of the multimeter and the other to the positive terminal.' Chemists often use the "red, right, plus" rule in connecting the red wire of the multimeter to the right-side positive electrode (cathode) of the galvanic cell 3. Determine the copper-zinc cell potential. If the multimeter reads a negative potential , reverse the connections to the electrodes. Read and record the (positive) cell potential. Identify the metal strips that serve as the cathode (positive terminal) Multimeter Cell Potential vs. pCu 1.150 1.100 1.050 1l 1.000 LU 0.950 0.900 0.850 0.800 0 2 8 6 4 1( pCu Figure 32,2 The variation of E,. 11 versus the pCu Figure 32,3 Setup for measuring the cell potentials of six galvanic cells 1 You have now combined two half-cells to form a galvanic cell. 360 Galvanic Cells, the Nernst Equation and the anode. Write an equation for the half-reaction occurring at each electrode. Combine the two half-reactions to write the equation for the cell reaction.<D 4. Repeat for the remaining cells. Determine the cell potentials for all possible galvanic cells that can be constructed from the four redox couples. Refer to the Report Sheet for the various galvanic cells. Prepare a new salt bridge for each galvanic cell.<%> 5. Determine the relative reduction potentials. Assuming the reduction potential of the Zn 2•(0. l M)/Zn redox couple is -0.79 V, calculate the reduction potentials of all other redox couples.2® [I] 6. Determine the reduction potential of the unknown redox couple. Place a 0.1 M solution and electrode obtained from your instructor in a small beaker. Determine the reduction potential, relative to the Zn2+(0. I M)/Zn redox couple, for your unknown redox couple.© 1. Effect of different molar concentrations. Set up the galvanic cell shown in Figure 32.5, using IM CuSO 4 and 0.001 M CuSO4 solutions. Immerse a polished copper electrode in each solution. Prepare a salt bridge (Part A.2) to connect the two half-cells. Measure the cell potential . Determine the anode and the cathode. Write an equation for the reaction occurring at each electrode.'2> B. Effect of Concentration Changes on Cell Potential 2. Effect of complex formation. Add 2-5 mL of 6 M NH 3 to the 0.001 M CuSO4 solution until any precipitate redissolves. 3 (Caution: Do not inhale NH 3.) Observe and record any changes in the half-cell and the cell potential.® 3. Effect of precipitate formation. Add 2-5 mL of 0.2 M Na 2S to the 0.001 M CuSO4 solution now containing the added NH 3• What is observed in the half-cell and what happens to the cell potential? Record your observations.CD 1. Prepare the diluted solutions. Prepare solutions I through 4 as shown in Figure 32.6 using a 1-mL pipet and 100-mL volumetric flasks.4 See Prelaboratory Assignment, question 3. Be sure to rinse the pipet with the more concentrated solution before making the transfer. Use deionized water for dilution to the mark in the volumetric flasks. Calculate the molar concentration of the Cu 2+ ion for each solution and record.® C. The Nernst Equation and an Unknown Concentration 2. Measure and calculate the cell potential for solution 4. Set up the experiment as shown in Figure 32.7, page 362, using small (-50 mL) beakers. Multimeter 1 M CuS04 0.001 M CuS0 4 Figure 32.5 Setup for measuring the cell Figure 32.4 A modern potential of a Cu 2• concentration cell multimeter Data Analysis, A l"". 1 ml 1 ml 1 ml i "'----0.1 M CuS04 Figure 32.6 Successive quantitative dilution, starting with 0.1 MCuSO 4 'Note: These are not standard reduction potentials because l M concentrations of cations at 25°C ore not used. 3 Copper ion forms a complex with ammonia: Cu 2•(aq) + 4 NH 3 (aq) -> [Cu(NH 3) 4] 2•(aq) 4 Share these prepared solutions with other chemists/groups of chemists in the laboratory. Experiment 32 361 Multimeter Cu Cu 2+ (4) so/Figure 32.7 Setup to measure the effect that Cu Cu (3) diluted solutions have on cell potentials [I] The Zn 2•rzn redox couple is the reference half-cell for this part of the experiment. Connect the two half-cells with a new salt bridge. Reset the multimeter to the lowest range (~200 mV). Connect the electrodes to the multimeter and record the potential difference, Ec,u. expt~ Calculate the theoretical cell potential Ec,u. calc· (Use a table of standard reduction potentials and the Nemst equation.)® 3. Measure and calculate the cell potentials for solutions 3 and 2. Repeat Part C.2 with solutions 3 and 2, respectively. A freshly prepared salt bridge is required for each cell. See data from Part A.3 for the potential of Solution 1. Data Analysis, F, G 4. Plot the data. Plot Ec,u. expt and Ec,u. calc (ordinate) versus pCu (abscissa) on the same piece of linear graph paper (page 368) or by using appropriate software for the four concentrations of CuS04 . Have your instructor approve your graph~ 5. Determine the concentration of the unknown. Obtain a CuS04 solution with an unknown copper ion concentration from your instructor and set up a like galvanic cell. Determine Ec,u as in Part C.2. Usii the graph, determine the unknown copper(II) ion concentration in the solution. [I] Disposal: Dispose of the waste zinc, copper, magnesium, and iron solutions in the Waste Metal Solutions container. Return the metals to appropriately marked containers. CLEANUP: Rinse the beakers twice with tap water and twice with deionized water. Discard the rinses in the Waste Metal Solutions container. The Next Step 362 Galvanic cells are the basis for the design of specific ion electrodes, electrodes that sense the relative concentration of a specific ion (e.g., hydrogen ion) relative to the electrode that has a fixed concentration. Part C of this experiment could be the apparatus for measuring concentrations of Cu2+ in other samples. According to equation 32.8, the pCu (negative log of [Cu 2+]) is proportional to the Ec,u! Research specific ion electrodes, their design, and their application. Design an experiment in which a specific ion electrode, other than the pH electrode, can be used to systematically study an ion of interest. Galvanic Cells, the Nernst Equation