ZrO2 Sensors for in Situ Measurement of pH in High-Temperature and

Mar 18, 2008 - hydrothermal fluids in hot springs vents at midocean ridge. These sensors, however, are not sensitive below. 200 °C. Zr/ZrO2 sensors h...
0 downloads 0 Views 143KB Size
Download PDF
Anal. Chem. 2008, 80, 2982-2987

Technical Notes

Zr/ZrO2 Sensors for in Situ Measurement of pH in High-Temperature and -Pressure Aqueous Solutions R.H. Zhang,* X.T. Zhang, and S.M. Hu

Open Research Laboratory of Geochemical Kinetics, Institute of Mineral Resources, Chinese Academy of Geological Sciences, Baiwanzhuang Road 26, Beijing 100037, P. R. China

The aim of this study is to develop new pH sensors that can be used to test and monitor hydrogen ion activity in hydrothermal conditions. A Zr/ZrO2 oxidation electrode is fabricated for in situ pH measurement of high-temperature aqueous solutions. This sensor responds rapidly and precisely to pH over a wide range of temperature and pressure. The Zr/ZrO2 electrode was made by oxidizing zirconium metal wire with Na2CO3 melt, which produced a thin film of ZrO2 on its surface. Thus, an oxidationreduction electrode was produced. The Zr/ZrO2 electrode has a good electrochemical stability over a wide range of pH in high-temperature aqueous solutions when used with a Ag/AgCl reference electrode. Measurements of the Zr/ZrO2 sensor potential against a Ag/AgCl reference electrode is shown to vary linearly with pH between temperatures 20 and 200 °C. The slope of the potential versus pH at high temperature is slightly below the theoretical value indicated by the Nernst equation; such deviation is attributed to the fact that the sensor is not strictly at equilibrium with the solution to be tested in a short period of time. The Zr/ZrO2 sensor can be calibrated over the conditions that exist in the natural deep-seawater. Our studies showed that the Zr/ZrO2 electrode is a suitable pH sensor for the hydrothermal systems at midocean ridge or other geothermal systems with the high-temperature environment. Yttria-stabilized zirconia sensors have also been used to investigate the pH of hydrothermal fluids in hot springs vents at midocean ridge. These sensors, however, are not sensitive below 200 °C. Zr/ZrO2 sensors have wider temperature range and can be severed as good alternative sensors for measuring the pH of hydrothermal fluids. Monitoring a hydrothermal reaction usually involves measuring the temperature, pressure, and activities of aqueous species and hydrogen ion. The pH is often the key variable that should be monitored and controlled, because those hydrothermal reactions often involve the hydrogen ion. The activity of the hydrogen ion * To whom correspondence should be addressed. Tel: +8610 68329535. Fax: +8610 68327063. E-mail: [email protected].

2982 Analytical Chemistry, Vol. 80, No. 8, April 15, 2008

(or hydroxide ion) affects the stability of the final product, and the pH may also affect the reaction rate. Recent studies of dissolution kinetics of mineral materials in aqueous solution have focused on the pH dependence of the reaction rates.1-4 The aim of this work was to develop the necessary techniques for measuring pH more quantitatively under extreme hydrothermal reaction conditions. Particularly, understanding the behavior of high-temperature aqueous solutions represents a new frontier in electrochemical studies that is both technically challenging and technologically important. In order to employ sensors for in situ measurement of the dissolved activity of H+ of hydrothermal vent fluids at midocean ridge and direct determination of aqueous solution chemistry under the extreme conditions, solid-state sensors were developed and used in the past.5-9 Recent advances in material science and sensor technology have resulted in the development of yttriastabilized zirconia (YSZ) ceramic based on pH, H2, and H2S sensors. Hg/HgO was put in a YSZ tube as an oxidationreduction electrode, i.e., YSZ/HgO/Hg. A YSZ/HgO/Hg electrode accompanied with AgCl/Ag was used to measure the pH of hydrothermal vent fluids at midocean ridge. This chemical sensor can be employed in high-temperature-pressure solutions up to 400 °C and 40 MPa.5-7 It is well-known that there is a large temperature gradient from the center of the hydrothermal vent to outside, from 405 to 2 °C. In the last 20 years, many works were reported on both measuring pH at elevated temperatures and pressures and (1) Casey, W. H.; Hochella, JKr. M. F.; Westrich, H. R. Geochim. Cosmichim. Acta 1993, 57, 785-793. (2) Mogollon, J. L.; Ganor, J.; Soler, J. M.; Lasaga, A. C. Am. J. Sci. 1996, 296, 729-765. (3) Oelkers, E. H. Geochim. Cosmochim. Acta 2001, 65, 21 3703-3719. (4) Zhang, R.; Hu, S. Aquat. Geochem. 2006, 12, 123-159. (5) Macdonald, D. D.; Hettiarachchi, S.; Lenhart, S. J. J. Solution Chem 1988, 17, 719-732. (6) Ding, K.; Seyfried, W. E. Science 1996, 272, 1634-1636. (7) Zhang, Z.; Ding, K.; Seyfried, W. E. 11th Annual V. M. Goldschmidt Conference, 2001; p 3391. (8) Zhang, X.; Zhang, R. High Technol. Lett. 2004, 10 (Suppl), 360-363. (9) Macdonald, D. D.; Liu, J.; Lee, D. J. Appl. Electrochem. 2004, 34, 577582. (10) Danielson, M. J.; Koski, O. H. J. Electrochem. Sci. Technol. 1985, 132, 296-301. (11) Niedrach, L. W. J. Electrochem. Sci. Technol. 1982, 129, 7 1446-1449. 10.1021/ac070684u CCC: $40.75

© 2008 American Chemical Society Published on Web 03/18/2008

interpreting electrochemical reactions on electrodes.5-6,8-17 Metal to metal oxide interfaces have been used to make an oxidation/ reduction electrode before.10-12 A metal/metal oxide couple that has the chemical stability and sensitivity in the pH ranges of interest under the expected measurement conditions (extreme conditions) should be selected.5-9 A number of metal/metal oxide electrodes (e.g., YSZ/HgO/Hg or Cu/CuO2, Ir/IrO2, Zr/ZrO2, W/WO3) have been examined at high temperatures up to 300 °C in the past several years, most notably by Macdonald and his group, and particularly the successful one is W/WO3.5,8-12 Nevertheless, previous attempts to employ these pH sensors in hydrothermal systems (e.g., at the midocean ridge) were partially successful at high temperatures above 200 °C. But YSZ/HgO/ Hg electrode is not sensitive at temperatures below 200 °C.7,8 A Ti/TiO2 electrode has been used to replace the YSZ/HgO/Hg sensing electrode to determine hydrothermal solution chemistry.7 Accordingly, we required a metal/metal oxide electrode of the pH sensor having one chemical valence of the metal oxidation state over a wide temperature range, which valence is not changeable particularly at high temperatures. Based on this principle, a suitable system appears to be Zr/ZrO2. It was found that Zr/ZrO2 has the same good nature as Ti/TiO2, with well-known corrosion resistance and favorable mechanical stability.8 Therefore, to assess the viability of the Zr/ZrO2 electrode (combined with the Ag/AgCl reference electrode) as a pH sensor in fluids with high salinity (3.5 wt % NaCl) and low to moderate pH, we performed experiments at temperatures up to 200 °C and pressures of 20-40 MPa. Furthermore, the probe of the pH sensor must be sensitive over a wide temperature range from 2 to 200 °C, and the pH sensor should possess corrosion resistance, chemical stability, and favorable mechanical stability under demanding reaction conditions. The work described in this paper proves the feasibility of performing pH measurements in hydrothermal environments at high temperatures and pressures. This work involved measuring the voltage of the Zr/ZrO2 sensor in a series of buffer solutions over a wide pH range and measuring the voltage of the sensor in the NaCl-HCl-H2O systems. The measured pH was compared with the theoretical expected pH to estimate the viability of the sensor systems. The experiments indicate that the Zr(Zr/ZrO2) sensor can be utilized in a wide temperature range as a good alternative to the YSZ sensor. Experiments also demonstrate that the Zr/ZrO2 electrode is a suitable pH sensor for monitoring hydrothermal systems at the midocean ridge or synthesis reactions at high temperatures of interest. The important innovation of the work described here is the selection of a Zr/ZrO2 pH sensor having an appropriate working (12) Kriksunov, L. B.; Macdonald, D. D. In Physical chemistry of aqueous systems: meeting the needs of industry; Proc 12th Int. Conf. Prop. Water and Stream; White H. L., Sengers J. V., Neumann D. B., Bellows J. C., Eds.; Begell House: New York, 1995; p 432. (13) Zhang, R.; Zhang, X.; Hu, S. Mater. Lett. 2006, 60, 3170-3174. (14) Macdonald, D. D.; Hettiarachchi, S.; Song, H.; Makela, K.; Emerson, R.; Ben-Haim, M. J. Solution Chem. 1992, 21, 849-881. (15) Kriksunov, L. B.; Macdonald, D. D. Sens. Actuators, B: Chem. 1994, 22, 201-204. (16) Macdonald, D. D.; Kriksunov, L. B. Electrochim. Acta 2001, 47, 775-790. (17) Lvov, S. N.; Palmer, D. A. In The Physical and Chemical Properties of Aqueous Systems at Elevated Temperatures and Pressures: Water, Steam and Hydrothermal Solutions; Palmer D. A., Fernandez-Prini, R., Harvey, A. H., Eds.; Elsevier Academic Press: Ansterdam, 2004; Chapter 11, p 377.

condition over a wide range of pH in high-temperature aqueous solutions, particularly at the midocean ridge.8 EXPERIMENTAL SECTION Melting of NaCO3 can favor oxidation of Zr to form ZrO2 thin film on the Zr metal surface. Zr metal wire (inner diameter ∼1 mm) with a purity of >98% was cleaned ultrasonically using acetone to remove fine particles at room temperature, rinsed with distilled water, and dried at 70-80 °C. Zr metal wire was put in a Na2CO3 melt in an Al2O3 crucible with an Au liner. Melting temperature is maintained at 890 °C for 1-1.5 h. ZrO2 thin film was formed on the surface of the Zr metal, and this method is used to make a Zr/ZrO2 oxidation/reduction electrode. The rest of the zirconium metal wire, except the sensor end (ZrO2 film on Zr) and the opposite end (connected to the voltmeter), was covered with a heat-shrinkable PTFE tube. The pH measurements under ambient conditions in standard buffer solutions were carried out with Zr/ZrO2 sensors. The Zr metal of Zr/ZrO2sensors had been oxidized in different time lengths, which showed a little different result. Only the melting oxidation of Zr metal for 1-1.5 h results in a good Zr/ZrO2 electrode, which can be employed in measuring the pH of high-temperature-pressure hydrothermal solutions. In this case, the 20-µm-thick ZrO2 film coated on Zr wire is solid and continuous. A Ag/AgCl electrode functioned as the reference electrode for the potential (pH) measurements. Put a silver wire in AgCl powders in the YSZ tube, and then melt AgCl. Thus, we found a thick AgCl film formed on the silver surface. And also, a ceramic porous plug was put in one end of the tube. This is the sensor part (Ag/AgCl) used for electrolytic contact with the test solution environment. A heat-shrinkable, PTFE tube was used to cover the other end of Ag wire, and sealed into the YSZ tube, from which part of the Ag wire was exposed for connection to the voltmeter. Standard buffer solutions were obtained from Hanna Instruments: HI7004 is the pH 4.01 buffer solution (25 °C, (0.01 pH), HI7007 the 7.01 buffer solution (25 °C, (0.01 pH), and HI7010 the 10.01 buffer solution (25°C ( 0.01 pH). NaCl-HCl-H2O (or NaCl-NaOH-H2O) solutions were also prepared with the source solution made of deionized and degassed water (conductivity grade water, Milli-Q system 18.2 MΩ cm), i.e., 3.5 wt % NaCl aqueous solution. A high-temperature and -pressure experimental system was set up to measure the Zr/ZrO2 sensor potentials. We performed experiments in a computer-controlled Ti flow reactor, which is especially resistant to corrosion by acid NaCl-bearing fluids. That is a flow-through reactor system with a pressure vessel, in which the chemical sensor can be put. The experimental flow reactor used for the high-temperaturepressure pH measurements is shown schematically in Figure la and b. A back pressure regulator was used to control the flowing fluid pressure in the measurement system. An additional pressure gauge provided pressure readings for the flowing system. Also, a pressure sensor is connected to the flow system, which provides an accurate pressure reading with (0.01 MPa (0.1 bar). A highpressure liquid pump offers a liquid flow velocity range from 0.1 to 9.9 mL/min. When we operated the measurement system of flowing fluids and maintained a constant high pressure of the Analytical Chemistry, Vol. 80, No. 8, April 15, 2008

2983

Figure 1. High temperature, high-pressure pH measurement system: (a) Measurement system consists of pressure vessel with multielectrode holes; furnace and temperature controller; pump, solution reservoir; back pressure regulator. (b) High-pressure vessel 1, Ag/AgCl electrode; 2, Zr/ZrO2 electrode; 3, fluid inlet and fluid outlet; 4, hole for thermocouple; 5, upper part of vessel; 6, lower part of vessel. c) Structure of Zr/ZrO2 pH sensor, 1, ZrO2 film on Zr thread; 2, mixed Teflon and graphite seal; 3, steel seal; 4, insulation coating and a heat-shrinkable PTFE tube; 5, high-pressure vessel; 6, pressure fitting; 7, Zr thread; 8, steel connect; 9, steel tube; 10, pressure fitting; 11, metal lead.

fluids, e.g., at 27 MPa, the pressure of the flowing fluids varied within (0.05 MPa (0.5 bar). The pressure of the measurement system was maintained to be stable through the back pressure regulator. The heating system was used to control the cell temperature, including a thermocouple, a heating furnace, and a proportional temperature controller. The temperature reading is accurate within (0.1 °C, and stability within (0.1 °C. There are seven holes made on the top of the pressure vessel, in which the fluid inlet, fluid outlet, thermocouple, and four different electrodes can be put. The thermocouple, tubing, Zr/ ZrO2 electrode, and Ag/AgCl reference electrode were sealed into the pressure vessel. The experimental system is connected to a computer, which can record all of the data of cell potentials, temperature, and pressure and monitor all of the data simultaneously. It can obtain six channels of data in 1 s. An advantage of this facility is that the Zr/ZrO2 electrode and associated reference electrode have direct access to the fluid phase at the experimental conditions, which ensure in situ measurements. Experiments commenced with the continuous flow (2-4 mL min-1) of 3.5 wt % NaCl solution having pH (25 °C) between 1.5 and 8. As prepared, the test solutions had varied pH of the source solution by adding HCl (adding NaOH for basic). As the fluids flowed onto the reactor, cell potential was measured with an electrometer with input impedance of 1013 Ω. The potential (pH) values for these test solutions at room temperature were measured using a Hanna pH meter and also measured using the Zr/ZrO2 sensor connected to our electrometer experimental system. The pH values for these solutions at the (18) Shvarov, Y. V. Geochem. Int. 1999, 6, 571-576.

2984

Analytical Chemistry, Vol. 80, No. 8, April 15, 2008

high temperatures were measured using the Zr/ZrO2 sensor connected to our electrometer experimental system and also calculated using an equilibrium solution model. Theoretical pH values of the test solutions are calculated using the HCh code, which equilibrium calculation and database are demonstrated to be identified.18 The deviation of cell potentials measured between using the Hanna pH meter and the Zr/ZrO2 pH sensor connected to our electrometer for measuring pH of the same solutions at room temperature is only 0.1 mV. The deviation in our measurements of cell potentials is generally within (0.5 mV at high temperatures. See the Appendix. RESULTS AND DISCUSSION This paper reports the first in situ measurement of pH using a Zr/ZrO2 sensor under high-temperature fluid conditions at the midocean ridge. The Zr/ZrO2 redox sensor was used to measure fluid pH at high temperatures and pressures combined with a Ag/ AgCl reference electrode. Thus, the overall electrochemical cell is as follows:

Ag|AgCl|Cl-, H+, H2O|ZrO2|Zr

Cell potential ∆E (V s)T,P as a function of pH can be described as follows:

∆E(V)T,P ) ∆E°cell +

2.3026RT [log a(Cl-1) - 1/2 log a F 2.3026RT (H2O)] pHT,P (1) F

Figure 2. Plot of Zr/ZrO2 sensor potential against measured pH at 20 °C in testing pH sensor at 20 °C.

where ∆E°cell is the cell potential at standard state, which can be calculated from the standard state potential of E°(Ag/AgCl)and E°(Zr/ZrO2); R is gas constant; F refers to Faraday constant; and a(Cl-1)and a(H2O) are the activities of Cl- and H2O, respectively. In the equation, T is K and pHT,P is pH at high temperatures and pressures. The sensor potential, ∆E(V)T,P is solely a function of pH at a given dissolved Cl and T-P condition. A series of measurement experiments of Zr/ZrO2 sensor were performed in NaCl fluids in the temperature range from 20 to 200 °C. The results revealed spectacular pH response (Figure 2). Thus, a plot of the Zr/ZrO2 sensor potential against the measured pH at 20 °C is shown in Figure 2. Figure 2 expresses that the measured cell potential varies linearly with pH of the fluids at 20 °C. This figure indicates that the line of cell potential against pH has the slope with -0.058 V/pH, and the correlation coefficient is -0.997. The intercept is ∆E°cell at 20 °C, which equals to 0.487 V. As predicted from theoretical considerations, pHT,P changed systematically with pH25°C. We calculated pHT,P corresponding to the measured pH25°C of the NaCl-HCl-H2O at 25 °C, shown in Figure 3. The measured cell potential at high temperatures and pressures that is directly related to pHT,P, (eq 1) can be compared with the theoretically predicted relationship from the revised Helgeson-Kirkaham-Flowers equation of state for aqueous species together with other appropriate thermodynamic data.19 Note that the HCh code (1999 version) utilizes the database of the SUPCRT92 version for theoretical calculation. These equilibrium models and the database for aqueous species and activity coefficients have been identical and have been applied to predict the relations between high-temperature electrochemical potentials and solution acidity.6 Measured cell potential values of the Zr/ZrO2 sensor at 200 °C, ∆ET,P against calculated pH at the higher temperatures and pressures, pHT,P are shown in Figure 4. Also, excellent linearity between ∆ET,P and pHT,P is observed at 200 °C. For a pH sensor at equilibrium, the Nernst equation predicts that the slope of the potential versus pH plot should be 2.303RT/ (19) Johnson, J. W.; Oelkers, E. H.; Helgeson, H. C. Comp. Geosci. 1992, 18, 899-947.

Figure 3. pH200°C as function of pH20°C in NaCl-HCl solution (3.5 wt %NaCl, and NaCl-NaOH-H2O for basic). Experimental measurements confirm the theoretically predicted relation between pH20°C and pH200°C, 27 MPa in a wide pH range.

Figure 4. Plot of Zr/ZrO2 sensor potential at 200 °C against pH200 °C, which was calculated from the measured pH of outlet solutions at 20 °C in testing sensor at 200 °C. Slope -97.89 mV; R2 ) 0.99.

F, where R equals 8.314 kJ K-1 kmol-1 and F equals 96 487 C mol-1, from which we obtained the theoretical values for the Nerstian slope of -58.1 and -94 mV at 20 °C and at 200 °C. Comparing with the theoretical slope values -58.1 (20 °C) and -94 mV (200 °C), we recorded the measured slope as -58.22 mV at 20 °C (R2 ) 0.997) and -97.89 mV at 200 °C (R2 ) 0.99). The excellent linear correlation between cell potential and pH and the Nerstian slope indicate that the cell-sensor response is a function of pH. The experimental slopes at 200 °C are slightly less than the theoretical values. Deviation from equilibrium theory is to be expected. The Zr/ZrO2 sensor in solution is not a strictly complete equilibrium system, but rather displays a mixed potential resulting from a balance between a partial anodic process (Zr + 2 H2O f ZrO2 + 4 H+ + 4e) and a partial cathodic reaction (e.g., H+ + e f l/2H2) occurring at the sensor surface.9,20 The electrochemical Analytical Chemistry, Vol. 80, No. 8, April 15, 2008

2985

instance, we must prepare pH buffer solutions. And, we need to monitor the response of cell potential and know when the cell becomes stable. Comparing previous research, this study reports a new preparation method for making a Zr/ZrO2 electrode by Na2CO3 melting and suggests a pressurized structure of electrode for monitoring high-temperature and high-pressure aqueous solutions (Figure 1c). Also, the electrochemical experiments demonstrate that the cell potential of Zr/ZrO2 responded to changes in the acidity of the pH variable source fluid with a wide pH range at temperatures from 20 to 200 °C. Comparing with theoretical Nernstian slope value -58.1 (20 °C) and -94 (200 °C), we recorded the measured slope as -58.22 mV (-0.058 22) at 20 °C (R2 ) 0.997) and -97.89 mV (-0.097 89) at 200 °C (R2 ) 0.99). The linear correlation between cell potential and pH and Nernstian slope indicates that the Zr/ZrO2 electrode responds consistently as a pH sensor. However, compliance with the Nernstian equation has been demonstrated in principle. The Zr/ZrO2 electrode can be exactly calibrated using a theoretical prediction of the NaClHCl-H2O solution. CONCLUSION

Figure 5. Measured voltage response of the Zr/ZrO2-Ag/AgCl cell to pH as demonstrated by flow-through experiment at 200 °C and 27 MPa. The recorded cell potential and cell temperature are plotted against time. (a) pH of input solution 3.5; (b) change pH of input solution to 1.9 after 4 h.

reaction occurring in the Zr/ZrO2 couple or ZrO2 tube, or YSZ tube, was reported before.14,21 Figure 5 shows cell temperature, pressure, and corresponding sensor potential data as a function of recording time over a period of 10 h. In spite of fluctuations in the pressure, the potential remains stable, and hence, the Zr/ZrO2 sensor is estimated to be viable for measuring pH in the NaCl-HCl-H2O system under the selected operating conditions. In the flow-through experiments at 200 °C, the sensor revealed rapid and reversible response to changes in pH of input aqueous solutions, shown in Figure 5. Figure 5a shows that changes in pH of input solutions from pH 3.5 to pH 1.9 were virtually instantaneously recorded and matched by corresponding changes in cell potential. This figure also notes that the potential of the Zr/ZrO2 electrode has a fluctuation with rising temperature from low temperature to 200 °C. It was found that the potential increased in first 20 min as temperature reached 200 °C, then decreased in the next 40 min, and finally it became progressively stable. This phenomenon proves that the ionic solid conductivity of a metal oxide is often affected by increasing temperature. If the electrode is used in field observation, we need to do calibration work; for (20) Kriksunov, L. B.; Macdonald, D. D.; Millett, P. J. J. Electrochem. Soc. 1994, 141, 3002-3005. (21) Chen, Y.; Urquidi-Macdonald, M.; Macdonald, D. D. Nucl. Mater. 2006, 348, 133-147.

2986 Analytical Chemistry, Vol. 80, No. 8, April 15, 2008

Through the above work, we could obtain conclusions as follows: (1) the experimental measurement of the Zr/ZrO2 sensor potential against a Ag/AgCl reference electrode indicates that the sensor potential varies linearly with pH of NaCl-HCl-H2O (and NaCl-NaOH-H2O) over a wide pH range at temperatures of 20 and 200 °C. The slope of the potential versus pH correlation at 200 °C is slightly lower than the theoretical value of the Nernst slope for the test solution at electrochemical equilibrium. This deviation is probably attributable to the fact that the sensor is not strictly at electrochemical equilibrium. (2) The Zr/ZrO2 sensors can be accurately calibrated under the high-temperature and -pressure conditions. Measurement of pH for the NaCl-HClH2O (and NaCl-NaOH-H2O) system is now possible in a wide temperature range, especially for the hydrothermal vent fluids at the midocean ridge without the limitations associated with available YSZ pH sensors. APPENDIX The pH standard buffer solutions were obtained from Hanna Ltd. Standard buffer solutions were prepared from Hanna ACS grade reagent and conductive grade water. The compositions of the solutions were 0.1 M NaOH, 0.5 M H3BO3 +0.2 M NaOH, 0.5 M H3BO3 + 0.05 M NaOH, 0.75 M H3BO3 + 0.02 M NaOH, and 1 M H3BO3. The pH values for these solutions at room temperature were measured using a Hanna pH meter. The pH of each solution at each of the test temperature was also calculated using an equilibrium model. ACKNOWLEDGMENT This study was supported by the Ministry of Science and Technology of China, Sea 863 project (2001AA612020-

3′2003AA612020-3), Basic Research Project (2001DEA30041, 2002DEA30084, 2003DEA 2C021), DY105-03-01, and NSFC

Received for review April 8, 2007. Accepted October 10, 2007.

(20373064, 50602042).

AC070684U

Analytical Chemistry, Vol. 80, No. 8, April 15, 2008

2987