Science

Submitted 3.29.06 | Revisions Received 6.28.06 | Accepted 7.11.06

The Influence of Reactant Concentration and Reaction Medium Ionic Strength, Viscosity, and Temperature on the Immunocomplex Substitution Reaction in the Radioimmunoassay of Aldosterone
Ricardo Dez Montoro, M. Teresa Salabert Salvador, Jose L. Moreno Frigols (Department of Physical Chemistry, Radioisotope Service, Valencia University Hospital, Valencia, Spain)
DOI: 10.1309/F3KCHH0QCMRLJECKV

Abstract
Background: Competitive protein binding radioimmunoassay (CPB-RIA) is a principal method for quantifying serum aldosterone concentration. The accuracy of this method is critically dependent on factors that influence the substitution reaction between unlabeled (Q) antigen (aldosterone) with 125I-labeled antigen (M) bound to anti-aldosterone antibody (P). We studied the influence of initial concentration of M, ionic strength, viscosity,

and temperature on the substitution reaction between M and Q. In addition, we propose a kinetic model for this reaction.

Methods: We used a commercially available CPB-RIA for aldosterone, a gamma counter, and a viscosimeter to study the effect of initial concentration of M, ionic strength, viscosity, and temperature on the substitution reaction between M and Q. Data were analyzed using Statistica software.

Results: The apparent rate constant for the reaction between M and Q in the formation of PM is dependent on the initial concentration of M, and the ionic strength, viscosity, and temperature of the reaction medium, and independent of the concentration of Q. Moreover, this reaction is endothermic and is not controlled by diffusion. Conclusion: Our results support the proposed model for the reaction between Q, M, and P: PM + Q PQ + M.

Competitive protein binding radioimmunoassay (CPBRIA) is used in quantitative measurement of serum aldosterone concentration. In this type of immunoassay,1 unlabeled antigen (Ag) present in the patients serum competes with radioactively-labeled antigen [Ag*, where * is typically iodine125 (125I)] for binding to an antibody (Ab) that reacts with both Ag and Ag* to form, after equilibrium is reached, the immune complexes, Ag-Ab and Ag*-Ab: Ag + Ab + Ag* (AgAb) + (Ag*-Ab). Because the concentration of Ag* and Ab are kept constant, the proportion of total immune complexes consisting of Ag*-Ab depends only on the amount of Ag (eg, aldosterone) in the sample being analyzed. If Ag* behaves similarly when bound and not bound to Ab, then separation of the bound and free fractions of Ag* is essential when quantifying the Ag concentration. Such a separation can be performed using a second Ab immobilized to the surface of a plastic test tube that recognizes an epitope on the first Ab, leaving only Ag-Ab and Ag*-Ab complexes in the tube after washing to remove any free Ag or Ag*. Therefore, the concentration of Ag*Ab complexes in the tube is dependent on the concentration of Ag in the patients serum. When that concentration is high, more Ag, and less Ag*, binds to the Ab, with formation of more Ag-Ab, and fewer Ag*-Ab, complexes, and lower counts from the radioactive label used in the assay. Similarly, the converse is true at low Ag concentrations. Therefore, in this type of immunoassay, the radioactivity measured is inversely proportional to the Ag concentration in the patients serum. The kinetics of these Ag-Ab reactions in reaching equilibrium are important determinants of the analytical sensitivity and accuracy of these immunoassay techniques in quantifying analyte (Ag) concentration.2-4 Moreover, such a diffusion-controlled reaction is affected by factors such as the temperature and viscosity of the reaction medium.
labmedicine.com

Previously, we proposed theoretical models for the process of immune complex formation in RIAs and immunoradiometric assays (IRMAs) and validated the applicability of these models, and their underlying mathematical equations, to the kinetics of the reaction occurring in RIAs and IRMAs.5-14 Currently, using a commercially available aldosterone RIA, we evaluated a new kinetic model for the substitution reaction between unlabeled antigen (Q; aldosterone) in the patients serum and labeled antigen (P; 125I-aldosterone) bound to antibody (M; antialdosterone) in the complex PM. In addition we studied the effect of diffusion, temperature, viscosity, and ionic strength of the reaction medium on the kinetics of this reaction.

Materials and Methods The theoretical model for the reaction between PM and Q that we studied is: PM + Q PQ + M anti-aldosterone antibody immobilized (coated) to the wall of a plastic tube M = 125I-labeled aldosterone PM = radioactive immunocomplex between P and M Q = unlabeled aldosterone PQ = non-radioactive immunocomplex between P and Q Control of the proposed reaction (ie, PM + Q PQ + M) by diffusion processes was studied using the equations (see Appendix I, page 61) developed and studied by Nygren15,16 and Stenberg17-19 containing 4 diffusion-associated variables. Moreover, a stepwise approach involving 4 stages was used to validate this reaction:
January 2007 Volume 38 Number 1 LABMEDICINE

P=

29

Science Obtain integrated rate equations for the overall processes. Study the effect of the temperature and viscosity of the reaction medium on the kinetics of the proposed reaction. To include or rule out the effect of electrical charges on the proposed reaction by studying the effect of the ionic strength of the reaction medium on the kinetics of this reaction. Demonstrate that the results from the studies performed in stages 1 through 3 above have applicability to improving the reaction conditions between PM and Q and to the design of immunoanalytical techniques based on this reaction. We obtained coated-tube RIA kits for aldosterone from Diagnostic Systems Laboratory (Webster, TX). These kits contained a solution of 125I-labeled aldosterone in a protein-based buffer; plastic tubes whose inside wall was coated with rabbit anti-aldosterone immunoglobulin; and, aldosterone standard solutions. Using glycerol and 2.05 M NaCl, we prepared a series of coated tubes containing varying amounts of 125I-aldosterone as shown in Table 1. All tubes (n=120) were incubated at room temperature overnight, followed by decanting the fluid and washing each tube with 2 mL of distilled water. In addition, we prepared 10 solutions containing different amounts of unlabeled aldosterone, glycerol, and NaCl as shown in Table 2. The viscosity of solutions 4, 5, 6, and 10 was measured at room temperature using a digital viscosimeter (Brookfield DVII, Brookfield Enginering Laboratories, Middleboro, MA). The viscosity of these solutions was obtained from a calibration curve prepared using viscosity measurements on calibrators containing differing proportions of glycerol and water. To study the influence of labeled and unlabeled aldosterone concentration and of ionic strength (I), viscosity (0), and temperature (T) of the reaction medium on the reaction between PM and Q, a series of 20 experiments were conducted using the coated tubes containing 125I-aldosterone (Table 1) and the solutions containing unlabeled aldosterone, glycerol, and/or NaCl (Table 2): Experiments 1-4: Study of the influence of 125I-aldosterone concentration (m) on the reaction between PM and Q using: Tubes 1-24 and solution 10. Experiments 5-8: Study of the influence of aldosterone concentration (q) on the reaction between PM and Q using: Tubes 25-48 and solutions 1, 2, 3, and 10. Experiments 9-12: Study of the influence of the ionic strength (I) of the reaction medium on the reaction between PM and Q using: Tubes 49-72 and solutions 7, 8, 9, and 10. Experiments 13-16: Study of the influence of the viscosity (0) of the reaction medium on the reaction between PM and Q using: Tubes 73-96 and solutions 4, 5, 6, and 10. Experiments 17-20: Study of the influence of temperature (T) using: Tubes 96-120 and solution 10. The baseline (t0) radioactivity in each tube was obtained by measuring the radioactivity in tube numbers 1, 7, 13, and 19 using a 1-min counting time and an ILKB Gammamaster Automatic Gamma Counter (Perkin Elmer, Turku, Finland). Reaction kinetics were studied by placing 1.0 mL of each solution required for each experiment into the appropriately numbered plastic coated tubes indicated above, followed by mixing, and incubation at room temperature for 12, 24, 48, 60, and 2,880 min. Incubation for 2,880 min (ie, 48 h) was considered adequate to represent reaction kinetics at infinite time. Each tube was washed to remove any unbound 125I-aldosterone. Radioactivity from 125I-aldosterone bound to the rabbit anti-aldosterone antibody coated wall of each tube was determined by counting each tube for 1 minute using a gamma counter.

Results and Discussion Values for radioactivity (y), initial concentrations (m) of PM immunocomplex and unlabeled aldosterone (q), ionic strength (I), viscosity (0), and temperature (T) are shown in Table 3. The conclusions from each group of experiments, relating to the proposed reaction model, PM + Q PQ + M, are summarized below and supported by the detailed mathematical analysis shown in Appendix II, pages 62-63:

Table 1_Preparation of Coated Tubesa Containing 125I-Aldosterone (M)

Coated-tube no. Reagent Added
125I-aldosterone

1-6 std, mL 0.25 0.75

7-12 0.50 0.50

13-18 0.75 0.25

19-120 1.00 0.00

H2O, mL
aPlastic

tubes whose interior wall was coated with rabbit anti-aldosterone antibody. std, standard; no., numbers

Table 2_Preparation of Solutions Containing 125I-Aldosterone (M)

Solution no. Reagent Added std, L Glycerol, L 2.05 M NaCl, L H2O, L
125I-aldosterone

1 25 0 100 7.875

2 50 0 100 7.850

3 75 0 100 7.825

4 100 1 100 6.8

5 100 2 100 5.8

6 100 3 100 4.8

7 100 0 200 7.7

8 100 0 300 7.6

9 100 0 400 7.5

10 800 0 800 62.4

Std, standard; no. number

30

LABMEDICINE Volume 38 Number 1 January 2007

labmedicine.com

Science Experiments 1-4: Influence of the initial concentration of the PM immunocomplex (m) Per Langmuirs model/equation, the20: initial activity of the radioactive immunocomplex (y0) is dependent on m apparent rate constant (kf) for the process is linearly dependent on the initial concentration of the radioactive immunocomplex. activity at equilibrium (ye) is directly proportional to y0, and therefore it also depends on m, and as a consequence of this finding, the RIA calibration curves obtained with these reagents must follow the 4 parameter logistic model and provide a good logit-log linear curve fit. Experiments 5-8: Influence of unlabeled aldosterone concentration (q) Equation [14] in Appendix II demonstrates that the kinetics of the reaction between PM and Q are independent of q. Thus, the q values we determined (Table 3) are lower than the analytical sensitivity of the aldosterone RIA used in this study. Experiments 9-12: Influence of ionic strength (I) of the reaction medium The effect of the ionic strength of the reaction medium on the kinetics of the reaction between PM and Q is appreciable. The reacting chemical species have electrical charges of an opposite sign (b3<0). Consequently, the reaction rate and concentration of PQ at equilibrium decrease at higher values for I. Experiments 13-16: Influence of viscosity (0) of the reaction medium An increase in the viscosity of the reaction medium causes the concentration of PQ at equilibrium to increase and the apparent rate constant to decrease.

Table 3_Summary Data for Experiments 1-20

Incubation time, min Exp no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 0 Tube no. 1 2,606 8 5,016 15 7,386 22 7,939 29 7,939 36 7,939 43 7,939 50 7,939 57 7,939 64 7,939 71 7,939 78 7,939 85 7,939 92 7,939 99 7,939 106 7,939 113 7,939 120 7,939 127 7,939 134 7,939 12 Y, cpm 2 2,040 9 3,042 16 4,661 23 5,433 30 5,920 37 5,321 44 5,139 51 5,433 58 5,433 65 5,107 72 5,233 79 5,395 86 5,433 93 5,082 100 4,897 107 5,453 114 5,433 121 7,384 128 7,354 135 7,725 24 3 1,823 10 3,072 17 4,333 24 4,258 31 4,355 38 4,873 45 4,343 52 4,258 59 4,258 66 3,732 73 3,981 80 4,059 87 4,258 94 4,475 101 4,015 108 4,173 115 4,258 122 6,169 129 7,278 136 7,743 36 4 1,421 11 2,651 18 3,418 25 3,950 32 3,716 39 4,275 46 3,975 53 3,950 60 3,950 67 3,296 74 3,424 81 3,443 88 3,950 95 3,919 102 3,266 109 3,457 116 3,950 123 6,141 130 7,148 137 7,394 48 5 1,252 12 2,594 19 3,151 26 3,923 33 3,706 40 4,043 47 3,430 54 3,923 61 3,922 68 3,522 75 3,436 82 3,368 89 3,923 96 3,451 103 2,696 110 2,804 117 3,923 124 5,748 131 7,016 138 7,287 60 6 1,254 13 2,553 20 3,176 27 3,562 34 3,384 41 3,867 48 3,165 55 3,562 62 3,562 69 2,356 76 3,321 83 3,169 90 3,562 97 3,158 104 2,527 111 2,307 118 3,562 125 5,449 132 7,049 139 6,810 2,448 () m, RU 7 1,101 14 2,086 21 3,296 28 3,693 35 3,259 42 3,971 49 3,736 56 3,693 63 3,693 70 2,422 77 3,444 84 3,080 91 3,693 98 2,346 105 1,790 112 1,095 119 3,693 126 4,000 133 4,159 140 4,777 25 50 75 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 q, RU 100 100 100 100 25 50 75 100 100 100 100 100 100 100 100 100 100 100 100 100 I, mol L-1 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 0.0513 0.0769 0.1026 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 0.0256 , mPa s 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.385 1.478 1.677 1.980 1.385 1.385 1.385 1.385 T,K 318 318 318 318 318 318 318 318 318 318 318 318 318 318 318 318 318 310 303 295

cpm, counts per minute (activity of PM immunocomplex); m, initial concentration [relative units (RU)] of PM; q, initial concentration (RU) of unlabeled aldosterone; I = ionic strength; , viscosity; T, temperature

labmedicine.com

January 2007 Volume 38 Number 1 LABMEDICINE

31

Science Experiments 17-20: Influence of temperature (T) of the reaction When temperature increases, the concentration of PQ at equilibrium and the apparent rate constant increase. From equation [19] and the Debye-Hckel, vant Hoff, and Eyring equations, the reaction is endothermic ()H0 = Rc = 8.311098 = 9124 Jmol-1) and the activation enthalpy ()H = Rh = 8.3122133 = 183925 Jmol-1) is much higher than the viscous flow energy of water. Therefore, the reaction is not diffusion-controlled. The greater the ability of the unlabeled antigen to displace the labeled antigen, the greater the analytical sensitivity of the RIA. Therefore, it is convenient to construct reaction conditions that favor the formation of non-radioactive immunocomplex (PQ). Based on our findings, this could be achieved by: Using low ionic strength solutions when preparing reagents; however, the ionic strength effect on the kinetics of the reaction is not too important and a decrease in the ionic strength or the reaction medium would slow the reaction. Increasing the viscosity of the reaction medium. This favors the formation of PQ but also at the expense of a slower reaction. Raising the temperature. The manufacturers instructions indicate that all testing using the DSL RIA kit for aldosterone should be performed at room temperature. However, the endothermic nature of the reaction between PM and Q favors considerably the formation of PQ at higher temperatures (eg, 37C), increases the rate of reaction, and provides results faster than incubation at room temperature.
1. Yalow RS, Berson SA. General aspects of radioimmunoassay procedures. In: In Vitro Procedures With Radioisotopes in Medicine. Vienna: IAEA, 1970, 455-481. 2. Zuber E, Mathis G, Flandrois JP. Homogeneous two-site immunometric assay kinetics as a theoretical tool for data analysis. Anal Biochem 1997;251:79-88. 3. Zuber E, Rosso L, Darbouret B, et al. A descriptive model for the kinetics of a homogeneous fluorometric immunoassay. J Immunoassay 1997;18:21-47. 4. Rabbany SY, Piervincenci RT, Kusterbeck AW, et al. Dissociation rate kinetics in a solid-phase flow immunoassay. Analytical Letters 1998;31:1663-1675. 5. Sadana A, Sii D. Binding kinetics of antigen by immobilized antibody: Influence of reaction order and external diffusional limitations. Biosens Bioelectron. 1992;7:559-568. 6. Olivas Arroyo C, Moreno Frigols JL. Influence of viscosity and ionic strength on the reaction kinetics of aldosterone and androstendione and their specific antibodies. J Pharm Biomed Anal. 2001;26:547-562. 7. Olivas Arroyo C, Duart Duart MJ, Moreno Frigols JL. Kinetics and equilibrium in insulin radioimmunoassay. J Immunoassay and Immunochemistry 2002;23:407-428. 8. Duart Duart MJ, Olivas Arroyo C, et al. Validation of a kinetic model for the reactions in RIA. Clin Chem Lab Med. 2002;40:1161-1167. 9. Garca Gmez J, Porcar Pons M, Moreno Frigols JL. Some kinetic aspects in the immunoradiometric assay of insulin-like growth factor binding protein-3. J Pharm Biomed Anal. 2002;29:307-315. 10. Garca Gmez J, Moreno Frigols JL. Kinetics and equilibrium in the immunoradiometric assay (IRMA) of thyroglobuline. J Immunoassay and Immunochemistry. 2002;23:347-367. 11. Garca Gmez J, Moreno Frigols JL. Comparison between mono- and biexponential models for reaction kinetics in the immunoradiometric assay of neuron-specific enolase. J Pharm Biomed Anal. 2003;33:891-901. 12. Garca Gmez J, Moreno Frigols JL. A two-site model for reaction kinetics in the immunoradiometric assay (IRMA) of skeletal alkaline phosphatase (sALP). J Immunoassay and Immunochemistry. 2007 (Accepted). 13. Garca Gmez J, Moreno Frigols JL. Kinetics aspects in the immunoradiometric assay (IRMA) of human interleukin-1 (Il-1). CAIJ. 2004;1:451-457. 14. Garca Gmez J, Moreno Frigols JL. Influence of concentrations, temperature, ionic strength, and viscosity in the immunoradiometric assay (IRMA) of parathyroid hormone (PTH). CAIJ. 2005;2:28-36. 15. Nygren H, Werthen M, Stenberg. M. Kinetics of antibody binding to solidphase-immobilised antigen. Effect of diffusion rate limitation and steric interaction. J Immunol Methods. 1988;101:63-71. 16. Nygren H, Stenberg M. Immunochemistry at interfaces. Immunology. 1989;66:321-327. 17. Stenberg M, Stiblert L. External diffusion in solid-phase immunoassays. J Theor Biol. 1986;120:129-140. 18. Stenberg M, Nygren H. Kinetics of antigen-antibody reactions at solid-liquid interfaces. J Inmunol Methods. 1988;113:3-15. 19. Stenberg M, Werthen M, Theander S, et al. A diffusion limited reaction theory for a microtiter plate assay. J Immunol Methods. 1998;112:23-29. 20. Barrow GM. Physical Chemistry Fourth Edition. Spanish Translation. Editorial Revert, 1988. 21. Atkins GW. Physical Chemistry Sixth Edition. Spanish Translation. Ediciones Omega, S.A., 1998, p.840. 22. Atkins GW. Physical Chemistry Sixth Edition. Spanish Translation. Ediciones Omega, S.A., 1998, pp. 227-229. 23. Atkins GW. Physical Chemistry Sixth Edition. Spanish Translation. Ediciones Omega, S.A., 1998, pp. 834-837.

Appendices continue on page 61

34

LABMEDICINE Volume 38 Number 1 January 2007

labmedicine.com

Science
continued from page 34

Appendix I The reaction studied:

PM + Q PQ + M

where, P = anti-aldosterone antibody immobilized on the wall of a plastic tube; M = 125I-aldosterone; PM = radioactive immunocomplex between P and M; Q = unlabled aldosterone; PQ = non-radioactive immunocomplex between P and Q The symbols used in the equations below: (PM)0 = w (Q)0 = q (PM) = w-x (Q) = q-x (PQ) = (M) = x k1= direct rate constant k2 = reverse rate constant Assuming that the reaction indicated above occurs in a single stage, the differential rate equation for this reaction is: dx/dt = k1(w-x)(q-x) k2x2 At equilibrium, the following equation must be true: 0 = k1(w-xe)(q-xe) k2xe2 From equations [1] and [2] we obtain: or dx/dt = (k1-k2)(xe-x)[(wqk1/(k1-k2)xe)-x] dx/dt = k(xe-x)(u-x) and, after integrating equation [4], Assuming that w>>q and q xe, and taking into account that: 1-(xe/u)exp(-(u-xe)kt) = 1-(xe2(k1-k2)/(wqk1))exp(-t(wk1+q(k2-k1)))) 1 Equation [4] becomes: x = xe(1-exp(-t (wk1+q(k2-k1)))) [5] Our experiments measured the radioactivity of the PM immunocomplex, represented by y, which is directly proportional to the concentration of PM. Therefore, the following equation is true: y0 /w = y/(w-x) = ye /(w-xe) = (y0-y)/x = (y0-ye)/ xe = From equations [5] and [6], we obtain: or and (y0-y) = (y0-ye)(1-exp(-t (y0kf+q(kr-kf)))) y = ye+ (y0-ye)(exp(-t (y0kf + q (kr-kf)))) y0kf + q(kr-kf) = the apparent rate constant for the reaction above [6] [7] [8] x = xe(1-exp(-(u-xe)kt))/(1-(xe/u)exp(-(u-xe)kt)) [3] [4] [2] [1]

(u-xe) k = (wqk1/(k1-k2)xe xe)(k1-k2) = (wk1/(k1-k2) q)(k1-k2) = wk1 + q(k2-k1)

labmedicine.com

January 2007 Volume 38 Number 1 LABMEDICINE

61

Science Appendix II Experiments 1-4: Influence of the initial concentration (m) of the PM immunocomplex Per Langmuirs model,20 the initial activity (y0) of the radioactive immunocomplex (PM) is dependent on m. In addition, the apparent rate constant (kf) for the reaction between PM and Q is linearly dependent on the initial concentration of the radioactive immunocomplex. Similarly, the activity at equilibrium (ye) is directly proportional to y0, and therefore it also depends on m. Consequently, the RIA calibration curves obtained with the reagents, P, M, and Q, obey a 4 parameter logit-log linear curve fit. When the results of experiments 1-4 are fitted to the equation: y = a1m/(m+b1) + (c1m/(m+b1) a1m/(m+b1))(exp(-t(d1m/(m+b1)+e1))) [9] where the parameters and coefficients for equation [9] are: a1 = 10103 b1 = 169.8 c1 = 22196 d1 = 0.0876 e1 = 0.0449 and r = 0.990, SS = 1.463106 Equation [9] was obtained from equation [7] by substitution. Assuming that the initially obtained y values are dependent on m, the following equation must be true: y0 = c1m/(m+b1) [10] Similarly, because y values at equilibrium are directly proportional to y0: ye = a1m/(m+b1) In addition, the equilibrium contant (K) for the reaction, PM + Q PQ + M, is: K = (PQ)(M)/(PM)(Q) Taking conservation of matter into account, the initial concentration of PM (PM0) is given by the equation: or (PM)0 = (PM) + (PQ) = (PM) + K(PM)(Q)/(M) = (PM)(1+K(Q)/(M)) (PM) = (PM)0/(1+K(Q)/(M)) Then, by introducing equations [6] and [11] into equation [10], and grouping constants, we have: ye = y0/(1+K(Q)/(M)) = [c1m/(m+b1)]/(1+K(Q)/(M)) = a1m/(m+b1) y0kf + q(kr-kf) = [c1m/(m+b1)]kf + q(kr-kf) = d1m/(m+b1)+e1 Experiments 5-8: Influence of unlabeled aldosterone concentration (q) The results of experiments 5-8 are fitted to the equation: y = a2 + (b2 a2)(exp(-t(c2 + d2q)) Its parameters and coefficient are: a2 = 3585 b2 = 7936 c2 = 0.0594 r = 0.988SS = 1.524106 AICc = 318.06 Equation [13] can be simplified to: Its parameters and coefficient are: y = a2 + (b2 a2)(exp(-t(c2))) a2 = 3582 r = 0.987 b2 = 7933 SS = 1.621106 c2 = 0.0688 AICc = 316.80 [14] [13] d2 = 0 .0001583 [12] [11]

Similarly, by introducing equation [10] into equation [8] and grouping the constants, we arrive at the final equation:

Equation [14] is obtained from equation [7] by substituting: ye = a2 y0 = b2 q(kr-kf) = c2 Equation [14] is preferable to equation [13], as it provides a lower AICc value. Experiments 9-12: Influence of ionic strength (I) of the reaction medium The results of experiments 9-12 are fitted to the equation: y = a3exp(b3I0.5)+(c3-a3exp(b3I0.5))exp(-t(d3exp(b3I0.5)+e3)) [15]

62

LABMEDICINE Volume 38 Number 1 January 2007

labmedicine.com

Science

Its parameters and coefficient are: b3 = -0.935 c3 = 7948 d3 = 0.0971 e3 = -0.00928 a3 = 3890 r = 0.980 SS = 3.012106 The rate constant is related to the ionic strength according to the Debye-Hckel equation21: k = k0exp(2.342zAzBI0.5) Equation [15] is obtained from equation [9] by assuming that m is negligible compared to b1, introducing the DebyeHckel expression in b1 and d1, and grouping the constants. Experiments 13-16: Influence of viscosity (0) of the reaction medium The results of experiments 13-16 are fitted to the equation: y = a4/(0+b4)+(c4- a4/(0+b4))exp(-t(d4/(0+b4)+e4)) Its parameters and coefficient are: a4 = 1214 b4 = -1.049 r = 0.993 SS = 1.400106 Using Kramers30 equation: k0/kv = A+B0/00 [17] where k0 and kv represent the rate constants corresponding to viscosities, equation [16] is obtained from equation [9] by assuming that m is negligible compared to b1, introducing Kramers equation [17] in b1 and d1, and grouping the constants. Experiments 17-20: Influence of temperature (T) of the reaction The results of experiments 17-20 are fitted to the equation: y = a5exp(b5/T) + (c5 - a5exp(b5/T))exp(-t(d5Texp(-e5/T)+f5)) Its parameters and coefficient are: a5 = 167.6 r = 0.995 b5 = 981 c5 = 7896 SS = 6.58105 d5 = 3.82105 e5 = 21397 f5 = 0.00412 [18] c4 = 7814 d4 = 0.01792 e4 = 0.013736 [16]

The equilibrium constant is related to the temperature according to the vant Hoff equation22: K = Aexp(-)H0/RT) The rate constant is related to the temperature according to Eyrings equation23: k = BTexp(-)H/RT) Equation [18] is obtained from equation [9] by assuming m is negligible compared to b1, introducing vant Hoffs equation in b1, Eyrings equation in d1, and grouping the constants. Experiments 1-20: Combined influences of m, q, I, 0, T The results of experiments 1-20 are fitted to the equation: Y = (am/(m+b))exp(c/T)(exp(dI0.5))/(0+e) + (fm/(m+b)-am/(m+b))(exp(c/T))/(0+e) exp(dI0.5))exp(-t((m/(m+b))gT(exp(-h/T))(exp(dI0.5))/(0+e)+j)) Its parameters and coefficient are: a = 166.6b =184.6 c = 1098 d = -1.687 e = -0.995 g = 5.481026 h = 22133 j = 0.00521 r = 0.990 SS = 9.571106 ye = (am/(m+b))exp(c/T)(exp(dI0.5))/(0+e) y0 = fm/(m+b) kr = (m/(m+b))gT(exp(-h/T))(exp(dI0.5))/(0+e) q(kr-k f) = j f = 22623 [19]

Equation [19] contains all of the previous equations and is obtained from equation [7] by substitution using the terms:

labmedicine.com

January 2007 Volume 38 Number 1 LABMEDICINE

63