KINETICS OF EXCHANGE OF AMMONIA1061
Vol. 4, No. 7, July 1965 infrared data, and suggests an energetic contribution to extraction by preserving the phenol dimer bonds. Acknowledgments.-The authors wish to express their appreciation to C. F. Coleman, K. B. Brown, and
C. F. Baes for their careful review and constructive criticism of the manuscript. We wish to thank W. E. Oxendine for the gas chromatograms and T . C. Rains and J. R. Lund of the ORNL Analytical Division for alkali metal and phenol analyses.
CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY, STATE UNIVERSITY, PULLMAN, WASHINGTON WASHINGTON
The Kinetics of Exchange of Ammonia with the Hexaamminecobalt(I1) Complex in Anhydrous Ammonia1 BY HANS H . GLAESER, HAROLD W. DODGEN,
AND
JOHN P. HUNT2
Received March 8, 1965 Kinetic data obtained by n.m.r. line broadening techniques are reported for the eo1'( NH3)e-NHa exchange reaction in liquid ammonia. The rate is essentially unaffected by addition of 1.7 m NaC104, 1.4 m NaC104 plus 0.25 m NH4C104, or 1.5 m HzO. Perchlorate and nitrate salts give similar results. The rate law R = 6kl[Co"(NH3)~] has been used. At -40.7' kl is 2.7 f 0.3 X lo4 sec.-l. The extrapolated value for k1 a t 25" is 7.2 & 1.4 X lo6 set.-'. Values for AH* and AS* (calculated from k l ) are 11.2 f 0.4 kcal./mole and 10.2 i 2 e.u., respectively. Some comparisons with Ni(I1) systems and other Co(I1) systems are made. The scalar coupling constant A / h was evaluated from the chemical shift and line broadening data and was found to be 7.5 f 0.1 X lo6 C.P.S.for CO"(NHS)G.
Introduction The research reported here is in continuation of a general program of study on ammonia exchange rates in transition metal complexes. Previous studies have been published concerning Cr(III),3Ni(II),4 and, less completely, Cu(II) systems. Experimental The n.m.r. techniques and equipment previously described in some detail4 have been employed. The ammonia was purified and solutions in ammonia were prepared and handled as described earlier.4 Cobalt reagents used were A.R. grade Co(N03)2*6H20, Co(C104)2.6 H 2 0 prepared from cobaltous carbonate and perchloric acid, and COIZprepared from cobaltous carbonate and hydriodic acid. The above salts were treated with anhydrous ammonia in the absence of air using a vacuum system. Upon repeated condensation of ammonia on the above salts followed by distillation of excess ammonia, violet anhydrous compounds were obtained which on the basis of Kjeldah1 analysis corresponded to the tetraammine salts. The iodide salt was found not soluble enough in ammonia for these studies (0.1 M ) over the range -70 to +30', giving pink solutions. The nitrate salt was soluble to the extent of >ca. 0.01 M a t temperatures below -20°, also giving pink solutions. Anhydrous sodium perchlorate and ammonium perchlorate were prepared as described b e f ~ r e . ~
mc0 refers
to the total molality of cobalt present in solution, y is the magnetogyric ratio for N14 (1934 gauss-' sec.-l), and A' is the line broadening (in gauss) due to Co(II), obtained from the full n.m.r. line widths a t half-maximum absorption using recorded absorption curves. The results are given in Table I. TABLE I LINE BROADENING DATAFOR Col'( NH3)B-NHIEXCHANGE T~~ )' x 105, T~~ x 105, t , OC.
A', gauss
m sec.
(A) Co = 0.050 ma 7.8 -21.5 0.65 0.36 14.2 -30.5
(B) -20.0 -36.3 -36.8 -40.0 -45.5
Co = 0.083 m 1.21 7.1 0.44 19.4 0.33 25.7b 0.26 32.4b 0.13 S4*
t, "C.
A', gauss
m sec.
( D ) Co = 0.25 m -36.3 1.29 20.1" -36.3 1.24 20.gd -36.3 1.15 22.7' -48.0 0.31 84" -49.8 0.28 93$ -50.2 0.21 124e ( E ) Co = 0.333 m 1 4 . 2 0.66 52 11.8 0.90 38 5 . 6 1.46 24.2 -35.0 1.92 17.9 -41.2 0.77 45 -43.2 61 0.56 60 -45.2 0.57 -51.2 0.27 127
(e) Co = 0 . 1 6 6 ~ ~ 25.8 0.10 172 - 3.1 1.27 13.5 11.9 - 6.8 1.45 -11.8 2.16 7.9 -23.8 2.21 7.8 -27.8 1.87 9.2 1.30 13.2 Treatment of Data and Results -35.0 64 -41.1 0.27 Our NH3 line broadening results are reported in a Co(N03)2.4NH3 used instead of Co(C104)~.4NHa. 0.01 m terms of the quantity Tz,," as in the Ni(I1) s t u d i e ~ . ~ C O ( N H ~ ) ~ ( Npresent. O~)~ 1.67 m NaC104 present. 1.42 m This quantity is defined as Tzp" 2mco/yA' where NaC104 plus 0.25 m NH4C104 present. e 1.52 m HzO present.
'
(1) This work supported in part by U.S.A.E.C. Contract AT(45-1)-1031. (2) To whom inquiries may be addressed (3) H. H . Glaeser and J. P. Hunt, Inorg. Ghem., 3, 1245 (1964). (4) H . H. Glaeser, G . A. Lo, H. W. Dodgen, and J. P. Hunt, i b i d , 4, 20G (1965). (5) R . Murray, H. W. Dodgen, and J. P. Hunt, ibid., 3, 1576 (1964).
A chemical shift (S) in the N14 absorption frequency in NH3 due to Co(I1) was observed. The shift data relative to pure NH3 are given in Table 11.
H. W. DODGEN, AND J. P. HUNT 1062 H . H. GLAESER, TABLE I1 CHEMICAL SHIFTDATA;mc0 s = (Uobad - W O ) / U O , t,
p.p.m.
O C .
25.8 - 3.2 - 6.8 -11.8 -23.8 5.6 Co = 0.333 172.
259 184 205 173 65 507
=
Inorganic Chemistry
0.166 II’S?7Zh.H8/llZC,
27.2 17.5 19.3 15.9 5.8 24,9“
We have modified our treatment5 of the data to include the expected temperature variation of the quantity Awg in the equation5 TzP” = r n ~ ~ p o / p 6 ~ 6 o A ~ 6 ~ lr??.Cop(,T6O/p6 (assuming only the species NHJ and CoI1(NH3)6 where pti is the atom fraction of NI4 in free ammonia, $6 is the atom fraction of N14 in CoI1(NH3)6,7 6 0 is the mean life (for exchange) of an ammonia ligand on the complex, and Awe is the n.m.r. absorption frequency (radians/sec.) for N14 in Coir ( N H J ) minus ~ the actual observed frequency in NH3 in the presence of the cobalt (we - u o b s d ) . I n the treatment of the data used here and previously we are also assuming that po ‘V 1 or that the solutions are dilute in paramagnetic species as was done by Swift and Connick.6 For our solutions “dilute” means p0 2 0.95. Under 3.2 3.4 3.6 4.2 4.4 4.6 lOa/T(’K.). these circumstances, A W 6 = ( W 6 - w,bsdj ‘V (06- ooj where WO is the absorption frequency for pure ammonia. Figure 1.-Plot of T2D”Tv s . 103/T and TSrnsH3/mCous. 103/T. For more concentrated solutions or more precise work the difference between w ~ I and , ~ ~w0 should be taken into culated from the TZP“data.j The straight lines in account in the Tz,” equation giving more cumbersome Figure 1 result from the fitting process. The lower expressions than those which follow. The temperaline on the left side (C) is drawn with the same slope ture dependence of TzP”can then be treated to a reaas the right-hand line (A). T o obtain the curve (B) sonable approximation as follonrs. The mean life, through the experimental points on the left side the 7 6 0 , is 6[CoI1(NH3)6]/K,lvhere R is the exchange rate points on C are multiplied by (T/TJ4 where Ti is the for NH3. Assuming R = Bk1[Co~~(NH3)61, 760 = l / k l temperature a t the intersection point. It may be where kl is the first-order specific rate constant for exnoted that use of data corresponding to line B only change of a particular ligand. without correction for variation of AWSwith temperaUsing the absolute reaction rate theory approach, ture leads to an appreciable error in the estimation of .ivelv;rite kl = (kT/h)e-AH*/RTeAS*/R . It is reasonable the kinetic parameters. The line through the shift to write that A w ~= a / T (Curie’s law). Making the points is calculated using lines A and B. The complete substitutions we obtain T?,,”= (mcOpo/p6)(kT3. curve for TZp”Tis obtained from the sum of A and B. e- AH */RTeAS*,’R ) / a 2 1 ~ (mCofiO/fi,3)e4H*’RT /(kT/h)e””*”. Slopes and intersection points were varied to give the The procedure to be followed a t this point will depend best fit for all the data. We calculate kl from the relaon whether the TZP” data lie largely (or are more pretion kl = 58.7/6Tz,”where 58.7 is the number of moles cise) in the region described by the first or by the second of ammonia per 1000 g. of ammonia. The value of term in T2p’’. If one divides TZ,” by T 3and plots log kl a t --110.7°is found to be 2.7 0.3 X I O 4 sec.-l. TzD”/T3 v s . 103/Tone can obtain - A H * from the limitThe extrapolated value for k1 a t 25” is 7.2 1.4X ing slope due to the first term (left-hand portion of 106 sec.-l. Using the slope of line 4 , and kl we calcurve) or multiplying 7‘2,’’ by T and plotting log T21,”T culate AH* = 11.2 =t0.4 kcal./mole and AS* = 10.2 US. 103/T one obtains AH* from the slope due to the + 2 e.u. At the intersection of A and B = 1/78,, = second term (right-hand portion of curve). In Figure 3S.7/GTZ,”. The scalar coupling constant A / h can 1 we have plotted log Gl1”Tand the quantity TSmNII,/ be obtained from the relation R / h = ( A w 6 / w ~ l z ) ; L k Y ‘ y / mco vs. 103/T. The shift ( S ) is defined by the relaS(S l ) y e = 7.4 X lo6 C.P.S. a t 25”. In this equation, tion S = ( u o b s d - wO)/’wO~ The shift data provide an A is the scalar coupling constant, h is Planck’s constant, independent criterion for fitting the TzP”data. The k is Boltzmann’s constant, uti the N1* resonance frequantity TSm~113/mcu falls to one-half its high-temquency in free ammonia, y the magnetogyric ratio for perature limiting value a t the intersection point of the N14, S the resultant electron spin for spin-free Co(I1) limiting TZP” curves, and the shift curve can be cal(1.5),and ye the magnetogyric ratio for the electron. AWOcan also be obtained from the limiting value a t (6) T. J. Swift and I
