- Email: [email protected]
Nickel(IV) dimethylglyoxime
434
JOURXAL
OF
ELECTRORSALYTICXL
B.ZCKEL(IV)
CHEXISTRY
DIBIETHYLGLYO-KIBIE
IKTRODUCTION
For many years the reaction of nickel(I1) d~ethylgl~~ox~e with an oxidizing agent in strong base to form a deep red colorl.” has been of use to analytical chemists both for spot tests and for calorimetric analysis. There has, however, been considerable disagreement in the literature as to the structure and form of the red colored material Some of the proposals include DHzNi(IV)Ol, Ni(III)DHaOHa, Ni(IV)DH20xa, Ni(IV)D$5.6, and an oxygen carrying chelate of unknown structure’. (For the sake of convenience dimethylglyoxime will be represented by DHZ.. DHzos stands for an oxidized form of dimethylglyoxime, probably furoxane.) In view of the possibility of the red material being an oxygen carrying chelate, it was decided that an electrochemical study to examine the reduction of oxygen attached to the metal chelate would be of interest_ Polarography, chronopotentiometry and controlled potential electrolysis were found to be of use. Magnetic susceptibility and spectrophotometry were also applied to this study. The results of this work, however, indicate that the red complex contains nickel in the +4 oxidation state and that all the evidence considered to support the idea of an oxygen carrying chelate cam be explained in these terms.
chemicals used in this study were reagent grade and used without further purification- Solutions of dimethylglyo_xime in 1.5 N sodium hydroxide were prepared fresh weekly. A n.ickel(II) nitrate stock solution was standardized by titration with EDTA. Polar&rams were recorded using a Sargent Model XV polarograph and a dropping mercury electrode for which nz = r-10 mg/sec and t = 7-10 set when h = 29.8 cm a_ndE = -o.4o V vs: the saturated calomel electrode. AII expeMrnents were carried out at 25.0 & o-x0. When desirable, solutions were deaerated by a rapid stream of nitrogen from which traces of oxygen were removed by a vanadous chloride scrubber. Controlled potential electrolysis experiments were carried out using an Analytical Instrumetits potentiostat and current integrator. The working electrode was of platinm gauze or merqrry. The cell construction for the platinum electrode has been previously described 8. The cell used with the mercury electrode was that supplied ~5th the_potentiostat. The values at which the electrode potential w& controlled iill
zlfICKEL(I\r)
435
DIMETHYLGLYOXIME
were selected by viewing polarograms or, in the case of the platinum electrode, chronopotentiograms taken at a shield platinum electrode in the usual mannerQ. Visible and U.V. adsorption spectra lvere recorded using either a Beckman DB or a DK-r spectrophotometer. Rlapetic susceptibility measurements were made employing the standard Gouy technique. The apparatus consisted of a Mettler single pan semi-micro balance, electromagnet, double end sample tube and heating coil for elevated temperature control. Measurements were made on solutions 0.002 M in Ni(N03)3, o.oogg M in dimethylglyoxime and 1.5 ~11 in NaOH at a field strength of 6500 gauss, and five different temperatures within the range 3gg”-34oOK. Since the measured weight differences were small, control measurements were run concurrently on solutions containing all components escept nickel and on a solution containing- only Ni(NO&, to ensure that the observed changes were due only to complesation. Assuming the validity of the additivity law, the susceptibility of the compIex was computed from the observed susceptibility differences between the complex and control solutions for which the concentrations of all constituents were known. The minimum accuracy in these determinations was established to be 1%. Data were taken both on solutions saturated with nitrogen and on solutions saturated with osygen. RESULTS
AND
DISCUSSIOS
The red o_xidized complex may be formed by dissolving solid nickel (II) .r~raro~~a~ily dimethyl~l~~o~ime in a solution of sodium hydra,xide and allowing oxygen from the air to contact the solution_ This is a slow process, however, and solutions are better prepared by adding a measured amount of nickel(II) solution to a dimethylglyoximesodium hydroxide solution Immediately upon the addition of nickel(I1). a yellow color forms -Ni(OH)a forms first but rapidly dissolves. If all the solution components are kept air free and no oxidizing agents were added, the reaction stops at this point. An absorption peak at 3So rnp was observed for such a solution. If the nickel dimethylglyoxime-sodium hydroxide solution was allowed to stand in contact with air (or treated with an oxidizing agent such as PbOs), the sofution @adually became dark red. (Absorption i;?timum 470 mp.) By mixing various amounts of nickel and dimethylglyoxime in x.5 iW sodium hydraxide it was found that a ratio of at least three moles of dimethylglyoxime per mole of nickel was necessary for ma-mum color intensity (and maximum diffusion current). A series of polarograms was then recorded with varying nickel concentrations between o-05 - 10-3 and I - x0-3 M. The concentration of dimethylglyoxime was 5 * 10-3 &f, and of sodium hydroxide 1.3 M. Various methods of oxidizing the solution to the red material were tried_ The most precise results were obtained by treating the solution with a srnal2 amount of PbOz for two hours. Bubbling pure oxygen for a comparable period was also successful, Air oxidation required several days and lead to somewhat erratic results due to the oxidation of the ~ethyl~lyo~me over this period of time. After PbOs or 0~ treatment the solution 1va.s placed in the polarographic cell, oxygen removed by passing nitrogen for IO r&n (a longer period allows some decomposition), and the polarograrn recorded. A typical polarogram is shown in Fig. I. The concentration range studied could not be extendedabove I.Io-aMdue to erratic behavior on the diffusion plateau. This is apparently due to the limited J.
E~ectronnat.
Chem..
8 (f964)
434-4-p
D.
436 solubility
of nickel(I1)
G.
D_%YIS,
E.
dimethylglyoxime
A.
BOUDREAUS
in 1-5 _M XaOH,
the erratic
ing from precipitation at or near the electrode surface. Extensive analyrical polarographic data was not gathered -tric methods’ would probably be superior in this concentration
since range.
behavior
result-
spectrophotomeHowever,
much
6.4 5.6
1.6
I
-02
I
-a3
I
I
-0.4 E,volts
Fig. 1_Polarogramofo.5-ro-3MnickelinasoIution Oxidized with Pb02, the escess of which
-0.5 vs.
I
I
1
-0.6 SC-E.
-07
-0.e
1.5 _M in NaOH and was removed by filtration was taken.
0.x AI in duncthylglyosimc. just before the polarogram
information was gamed from the polarographic study. The diffusion current constant, I, was found to have a value of 2.4 + 0.1. This value of I is in agreement with that of 2.4 for lead tartrate complex (a presumably similar ion) in sodium hydroxide and thus indicates a 2 electron reduction_ That the reduction controlled was shown by the fact that ;/r/J E was constant to z%_ rial was-found to be -o_zg7 V ZIS. S.C.E_ and was independent
current was diffusion The half wave potenof the nickel and di-
_methylglyoxime concentrations over at least a ten fold rauge. Thus the reduction is probabl$ reversible and the same number of dimethylglyoxime ligands are associated with the oxidized (red) form and the mckel(IIj complex. A plot of log i/(&-i) ZJS.E yielded a straight line with a slope of 0.030, confirming a reversible z electron reduction_ On the basis of the above data the electrode reaction would appear to be hTi(IV)Dr
+
zz e Z+ Ni(Ii)D,
where x is 3 or more_ Actually no statement can be made ide-ion participates ‘m the electrode reaction. Variations
as to whether of hydroxide
/_ Elecivoanal.
Chem..
or not hydroxconcentration 8 (IgG4) 434-441
xICKEL(IV)
were
not
attempted
since
a large
DIMETHYLGLYOXIME
enough
change
437
to result
in detectable
half-wave
potential shifts could not be accomplished without making the red complex unstable. Although the polarographic data strongly indicates CoiatroZZed potential electrol~~sis a two electron reduction, further tests were desired to confirm the fact that a two electron reduction was indeed occurring and that the reduction was between Ni(IV) and Xi(II),
and did not involve
the reduction
of molecular
oxygen
bound
to the nickel
as suggested
in ref. 7_ with a mercury pool electrode, was run at X controlled potential electrolysis, -0-60 V us. S-C-E. This potential is reducing enough to reduce the red complex (see dimethylglyoxime or hydrogen peroxide Fig. I) but not so reducin g that nickel(I1) (which might result from the reduction of bound molecular oxygen) would react. Thesolutionscontained o.~~-~o-s.M of nickel(I1) and IO-IO-3Mof dimethylglyoximein IOO ml of 1.5 M NaOH. Complete osidation was accomplished by treatment with PbO-) for five hours_ The excess solid PbOz was removed by filtration just before the electrolysis_ Dissolved osygen was removed by a stream of nitrogen. The average gz value for the controlled potential reductions at the mercury cathode was 1-9. This value fell slightly short of 2.0 due to electrode fouling by nickel(I1) dimethylglyoxime precipitating near the end of the electrolysis and possibly to a decomposition of the red nickel(W) dimethylglyoxime complex. The solution resulting from the electrolysi- s was light yellow and had the same spectra as the nickel(I1) dimethylglyo_time solution made by mixing the air-free components. A “titanium spot test” for hydrogen pero-xide was negative_ By recording current-rel-ersal chronopotentiograms using a platinum working electrode it was found that air-free nickel(I1) dimethylglyoxime could be o_xidized V vs_ SCE. and a reverse (reduction) wave (having a transition time at E* = f-o.22 one-third that of the oxidation transition time) with an E+ = -0.34 V vs. S.C.E. could be recorded. The opposite case could also be accomplished, starting with nickel(IV)dimethylglyoxime. When the current was varied, the quantity z& was found to be constant. It may thus be concluded that the electrode reaction at platinum is diffusion controlled and of the simple irreversible typelo. On the basis of the above values of Ea, a controlled potential electrolysis study was initiated nickel(I1)
at a platinum electrode. An air-free solution containing O.O~OO-IO-~ M and 0.3-10-3 M of dimethylglyoxime in r.5 M sodium hydroxide was made
of up
under nitrogen in the controlled potential electrolysis cell. The passage of nitrogen was continued throughout the experiment. The solution was oxidized electrolytically at a controlled potential of to.40 V ‘us. S.C.E. The deep red color of the oxidized nickel dimethylglyoxirne complex rapidly appeared, indicating oxygen or oxygen producing species to be unnecessary_ The electrolysis current eventually dropped to a low but finite value (direct oxidation of dimethylglyoxime)_ After correction for this background current an 77 value of x.92 was calculated. The working electrode was now made the cathode, the controlled reduction potential being -0.40 V zs. S.C.E. As the electrolysis proceeded the red color faded to the original yellow. The IZ value for reduction was found to be r-99. No background correction was necessary. Since the oxidation and subsequent reduction of nickel in a solution of dimethylglyoxirne-r-5 M sodium hydrouide’ showed analytical promise for a range of nickel concentrations generally too high for spectrophotometric work, an accuracy and interference study was made. The electrolyte, the concentrations of which were 1.5 M J_ Ebclroanal.
Chem..
8 (1964)
434-441
ID. G.
435
DAVIS,
E.
A.
BOUDREAUX
in sodium hydroxide and 0.01 1M in dimethylglyo_xime, was placed in the cell. A nickel sample was added, some of which was oxidized by dissolved air. The oxidation was completed by controlled potential electrolysis at to.30 V z~rs. S C.E. The quantity of electricity passed was not recorded. TVhen the current dropped to a low constant value, the electrolysis was stopped and oxygen removed by bubbling nitrogen through the solution_ The oxidized nickel was then reduced at -0-40 V vs. S.C.E. and the coulombs used to accomplish this, recorded. Table I indicates the results obtained TABLE THE
COMTROLLED
POTESTIXL
1 DETERS!IS.~TIOX
OF
NICKEL
.-L&fed ion {trreqtriv.) ---
Nicket takem fmeqzris-)
A-zckelfozrnd fi??.eptV.)
O.OQ8Q
o.ogS3 o.ogS6
o-0500
a_ogos 0.0~00 o-0499 a-0499
Co’+
0.0622
bln”‘(o.02)
0.04s3
Fe=
o.ozSS
IF&+ (o.o=rr)
0.0296 0.0301
co’” x3*
O-0303
al’”
(o_a3) (0.03)
0_03az
ci-
(o.oc))
0_0300
(a_og) (o.02) (0.05)
and the effect of interferences. In general, results can be obtained with an accuraqof better than 0_5~~~_Manganese and iron appear to interfere, the latter apparently due to electrode fouling which causes incomplete o,xidation of the nickel. The controlled-potential electrolysis evidence seems to substantiate the idea that the red-colored material is, indeed, a nickel(IV) dimethylglyoxime complex, rather than one of nickel(.KII) or invoIving an oxygen carrying chelate of nickel(U). Some additional evidence was found through spectrophotometry and magnetic measure-
rixents.
of the main points in favor of the idea that the red colored complex was of the oxygen carrying chelate variety was that the color could be changed to light yellow by heating to about 70” or holding the solution under partial vacuum for a few minutes’_ At first thought, it would seem that these procedures remove oxygen from solution and thus reverse a reversible oxygen chelate formation_ -Indeed; if nitrogen is passed through a previousIy oxidized nickel dimethylglyoxime solution, the color also fades slowly (and the polarographic wave of Fig. I disappears). -In view of the controlled potential electrolysis experiments and the fact that dimethylglyoxime is oxidizable at a platinum electrode at potentials slightly more oxid&ing t&n necessary for the formation of nickel(W). it appears that the nickel(IV) dimetbylglyo~e complex is unstable. The nickel(IV) is a strong enough oxXizing agent to i3xidize dimethylglyoxime slowly, probably to _furoxane6. This would -&.c-explain the disappearance of the red color on heating, etc. Toir&stigate this reaction more carefully a sample of nickel(N) in excess oxygenSpi?Xt~OpJzOtOF?Z&-
One
J_ EZ~cfman~Z. Chem_. 8 frg64)
a$$-+#1
KICKEL(IV)
DIMETHYLGLYOXIME
439
free dimethylglyoxime solution, 1-5 M in sodium hydro.xide, was sealed in a spectrophotometer cell and spectra taken over a period of several days (Fig. 2). The final spectra is identical with that of unoxidized nickel(I1) dimethylglyoxime. The fact that the comples changes spontaneously to the yellow nickel(I.1) complex in a
I
320
t
360
I
400
I
440
I
400
Wovelength
I
5;o
560
I
600
.mp
t
640
6
0
spectra of 24.0- 10-5 ~11nickel in a solution 1-5 AT in Ku’aOH and .+- 10-1 .M in The solution was treated with oxygen. the e_xcess osygen removed with sealed in the absorption cell Curve A, taken immedlatel>-; curt-e 33. after Z+ h; curare C,aftcr 85 h (no further change observed.)
Fig. 2. Absorption dimethylgl~oxirne.
nitrogen
and
sealed vessel at room temperature indicates that the instability is only accelerated by heat rather than being caused by it. The application of vacuum simply removes dissolved oxygen which othenvise would continue to oxidize the nickel. This evidence again supports the idea that the red complex may well be a nickel(IV) species-the nickel(W) being a strong enough oxidizing agent to react with dimethy~glyoxime. Ox&Z AND &RNEK" have shown by means of photometric titrations that the red complex has a ratio of three dimethylglyoxime ions for each nickel ion, and by means of paper electrophoresis that the charge on the red complex is -2 _ Experiments in this laboratory tend to confirm these results. In the light of this work and the polarographic studies, the electrode reaction at mercury and probably platinum electrodes can be written
The nickel(IV)
complex
can then decompose
slowly at room temperature
where DO= is most likely furoxane. (An attempt to isolate furoxane was unsuccessful but it is a known oxidation product of dimethylglyoximell). Magnetic data In an effort to establish definitely that the oxidation state of nickel in the red complex was indeed +4, a magnetic spsceptibtity study was conducted. J_ Ekiroanal.
Chews.. 8 (1964)
434-441
440
D.
G.
D_4VIS,
E.
A.
BOGDREAGX
The results of these deterlminations yielded for the complex effective molar susceptibilities, X;?, of +640 . IO-~ and +-64s - 10-6 for nitrogen and oxygen saturated solutions respectively. (Solutions saturated in osygen were corrected for the susceptibility of the dissoLved oxygen) The magnetic moment calculated from these data is r.23 =f= o.oz B.M., assumin g a simple Curie Law behavior. The temperature variation within the range 2gg0studies showed a change in susceptibility of 2.4 -F_ 1.0% 33g”K, whereas the expected temperature change (assuming the Curie temperature relationship) for a paramagnetic ion with one unpaired electron should have been at least 4% within the limits of experimental error. Thus the observed temperature variation is just within the limits of experimental error, while that expected for one unpaired electron is a factor of three times greater than what could be attributed to experimental error. Hence we are forced to conclude on the basis of these limited studies that the susceptibility of Ni(IV)Da-2 is independent of temperature, at least within the restricted tenperature range of investigation and high dilution conditions_ Although the value determined for the magnetic moment at ambient temperature should be reasonably accurate, too much significance should not be attached, however, to the temperature variation studies If the susceptibility is indeed temperature independent than the susceptibility must be of the high frequency type, wherein the paramagnetism arises from mixing of a singlet ground state with low lying singlet excited states z&z orbital angular moments. However, the observed magneticmomentis too large by about a factor of two to account for this effect alone. Magnetic moments of the magnitude found here have also been reported for other with Ni(IV) compoundsls as well as for Ni(I1) complexes in planar four-coordination a diamagnetic ground state 13. In the latter cases, if the Ni(IT) ion is in a weak tetragonal field then the paramagnetism should arise from mixing of the singlet ground stat& with a thermally accessible triplet state of the complexl*. Although alternative mechanisms may be equally valid for explaining the anomalous magnetic behavior of four-coordinated Ni(IT) complexes 13, the thermal population of a tnplet level is the most reasonable explanation for the anomalous magnetic behavior of a six-coordinated Ni(IV) complex_ That ~e~y~glyo~e is a strong field @and is evidenced by the fact that Ni(II)Dz, an effectively- square planar complex, is diamagnetic. Since the effective ligand field stieng$h increases proportionally with increasing positive charge on the central metal atom, then octahedrally coordinated Ni(IV)Ds s- should certainly e,xist in a diamagnetic singlet ground state. Consequently, in addition to high frequency effects, the param &gnetism observed in this Iatter complex should result from the population of a thermally accessable triplet level. If this interpretation is correct then the magnetic moment should increase with temperature, which has not yet been compl+ely confirmed for the complex in solution. Future investigations will include the attempt to isolate the complex and subject it to criti&I- magnetic and spectral investigations, Theoretical correlation with the experimentalobservables will also be attempted in an effort to gain a better under~standing of the bonding- ih this complex_ ACKNOWLEDGEMENTS
iavesrigation from the N&bnal
This
We sr.Qported in part by Public H&a&h Service Grant AM08z48 Instituttit of Arthritis and_ Metabolic D_iseases. J_
EZtztvounaL
Cliem.,‘El
(Ig6q)
43=9-+4x
s;ICKEL(IV)
44=
DIMETHYLGLYOXIME SUMMARY
The
red solutions
of nickel
dimethylglyoxime
in strongly
basic
solution
have
been
studied by electrochemical, spectrophotometric and magnetic methods. It is concluded that the red complex is a species of Ni(IV). A controlled potential coulometric method for the analysis of nickel is also presented. REFERENCES I F. FEIGL, Ber.. 57 (1924) 75S_ 2 G HAIM AID l3. TXRRANT, Ind. Etzg. Clrertr., IS (lg@) 51. AcLa Chettr. Scnnd., 3 (1949) 497. @I. 3 I<_ Ax JASSEN AND EL XYGA_XRD, 4 hl. HOORE>IAN, _4maZ. Chim. Ada, 3 (1949) 635. =j A. OK&? AND &I. SIR-VEK, CO~~C~~O?Z CxCh. Chel?a.Comm~m_, 24 (1959) 1-699. Ed B. JEZOWSK_~-TRZEBI_~TOWSK_~, 6 -4. OI-&Z, Theovy alrd StrmLzcve oj Complex Compourz~, Pergamon Press Inc.. New York. 1964. p- 167~ 7 J_ SELBINAKD J_ H_ Jus~~rv.J..~m. CAenz. SOL. Sz (rgGo) 1057_ 8 D. G. D_4vrs, J_ EZecrroaxaZ. Clretn., I (1959) 73_ g D G. DAVIS. =Inal. Chem.. 33 (IgGI) 1839. IO W. REIN~IUTH. rival. C7lenz.,32 (IgGo) rgr+ Ed. L. MEITLS. BlcGraw-Hill Book Co. II I?. C. &I_4RKEx_4S, Hat&book of AtzaZyticaZ Chettzislry. Inc., I\l;ewYork, 1963, section I?-,p- 14'. I?- I?.ROY _4x~ B. S_~R~I,I~, J_ I?zdiau Chenr. SOL, z=j (rg+S) zo=j_ 13 S. L.HoLT, R. J. BOUCHARD ASD R-L. CXRLI~, J. Am.Chenr. Sot., SG (IgGa) 519. 14 C. J_ B_~LLH=~USE~ .4i‘;~ X D_ LIEHR. J_ Am. Chetn. Sot., SI (1939) 535. J_ Eleclroatzal.
Chettz.,
8 (IgGq)
434-441
JOURXAL
OF
ELECTRORSALYTICXL
B.ZCKEL(IV)
CHEXISTRY
DIBIETHYLGLYO-KIBIE
IKTRODUCTION
For many years the reaction of nickel(I1) d~ethylgl~~ox~e with an oxidizing agent in strong base to form a deep red colorl.” has been of use to analytical chemists both for spot tests and for calorimetric analysis. There has, however, been considerable disagreement in the literature as to the structure and form of the red colored material Some of the proposals include DHzNi(IV)Ol, Ni(III)DHaOHa, Ni(IV)DH20xa, Ni(IV)D$5.6, and an oxygen carrying chelate of unknown structure’. (For the sake of convenience dimethylglyoxime will be represented by DHZ.. DHzos stands for an oxidized form of dimethylglyoxime, probably furoxane.) In view of the possibility of the red material being an oxygen carrying chelate, it was decided that an electrochemical study to examine the reduction of oxygen attached to the metal chelate would be of interest_ Polarography, chronopotentiometry and controlled potential electrolysis were found to be of use. Magnetic susceptibility and spectrophotometry were also applied to this study. The results of this work, however, indicate that the red complex contains nickel in the +4 oxidation state and that all the evidence considered to support the idea of an oxygen carrying chelate cam be explained in these terms.
chemicals used in this study were reagent grade and used without further purification- Solutions of dimethylglyo_xime in 1.5 N sodium hydroxide were prepared fresh weekly. A n.ickel(II) nitrate stock solution was standardized by titration with EDTA. Polar&rams were recorded using a Sargent Model XV polarograph and a dropping mercury electrode for which nz = r-10 mg/sec and t = 7-10 set when h = 29.8 cm a_ndE = -o.4o V vs: the saturated calomel electrode. AII expeMrnents were carried out at 25.0 & o-x0. When desirable, solutions were deaerated by a rapid stream of nitrogen from which traces of oxygen were removed by a vanadous chloride scrubber. Controlled potential electrolysis experiments were carried out using an Analytical Instrumetits potentiostat and current integrator. The working electrode was of platinm gauze or merqrry. The cell construction for the platinum electrode has been previously described 8. The cell used with the mercury electrode was that supplied ~5th the_potentiostat. The values at which the electrode potential w& controlled iill
zlfICKEL(I\r)
435
DIMETHYLGLYOXIME
were selected by viewing polarograms or, in the case of the platinum electrode, chronopotentiograms taken at a shield platinum electrode in the usual mannerQ. Visible and U.V. adsorption spectra lvere recorded using either a Beckman DB or a DK-r spectrophotometer. Rlapetic susceptibility measurements were made employing the standard Gouy technique. The apparatus consisted of a Mettler single pan semi-micro balance, electromagnet, double end sample tube and heating coil for elevated temperature control. Measurements were made on solutions 0.002 M in Ni(N03)3, o.oogg M in dimethylglyoxime and 1.5 ~11 in NaOH at a field strength of 6500 gauss, and five different temperatures within the range 3gg”-34oOK. Since the measured weight differences were small, control measurements were run concurrently on solutions containing all components escept nickel and on a solution containing- only Ni(NO&, to ensure that the observed changes were due only to complesation. Assuming the validity of the additivity law, the susceptibility of the compIex was computed from the observed susceptibility differences between the complex and control solutions for which the concentrations of all constituents were known. The minimum accuracy in these determinations was established to be 1%. Data were taken both on solutions saturated with nitrogen and on solutions saturated with osygen. RESULTS
AND
DISCUSSIOS
The red o_xidized complex may be formed by dissolving solid nickel (II) .r~raro~~a~ily dimethyl~l~~o~ime in a solution of sodium hydra,xide and allowing oxygen from the air to contact the solution_ This is a slow process, however, and solutions are better prepared by adding a measured amount of nickel(II) solution to a dimethylglyoximesodium hydroxide solution Immediately upon the addition of nickel(I1). a yellow color forms -Ni(OH)a forms first but rapidly dissolves. If all the solution components are kept air free and no oxidizing agents were added, the reaction stops at this point. An absorption peak at 3So rnp was observed for such a solution. If the nickel dimethylglyoxime-sodium hydroxide solution was allowed to stand in contact with air (or treated with an oxidizing agent such as PbOs), the sofution @adually became dark red. (Absorption i;?timum 470 mp.) By mixing various amounts of nickel and dimethylglyoxime in x.5 iW sodium hydraxide it was found that a ratio of at least three moles of dimethylglyoxime per mole of nickel was necessary for ma-mum color intensity (and maximum diffusion current). A series of polarograms was then recorded with varying nickel concentrations between o-05 - 10-3 and I - x0-3 M. The concentration of dimethylglyoxime was 5 * 10-3 &f, and of sodium hydroxide 1.3 M. Various methods of oxidizing the solution to the red material were tried_ The most precise results were obtained by treating the solution with a srnal2 amount of PbOz for two hours. Bubbling pure oxygen for a comparable period was also successful, Air oxidation required several days and lead to somewhat erratic results due to the oxidation of the ~ethyl~lyo~me over this period of time. After PbOs or 0~ treatment the solution 1va.s placed in the polarographic cell, oxygen removed by passing nitrogen for IO r&n (a longer period allows some decomposition), and the polarograrn recorded. A typical polarogram is shown in Fig. I. The concentration range studied could not be extendedabove I.Io-aMdue to erratic behavior on the diffusion plateau. This is apparently due to the limited J.
E~ectronnat.
Chem..
8 (f964)
434-4-p
D.
436 solubility
of nickel(I1)
G.
D_%YIS,
E.
dimethylglyoxime
A.
BOUDREAUS
in 1-5 _M XaOH,
the erratic
ing from precipitation at or near the electrode surface. Extensive analyrical polarographic data was not gathered -tric methods’ would probably be superior in this concentration
since range.
behavior
result-
spectrophotomeHowever,
much
6.4 5.6
1.6
I
-02
I
-a3
I
I
-0.4 E,volts
Fig. 1_Polarogramofo.5-ro-3MnickelinasoIution Oxidized with Pb02, the escess of which
-0.5 vs.
I
I
1
-0.6 SC-E.
-07
-0.e
1.5 _M in NaOH and was removed by filtration was taken.
0.x AI in duncthylglyosimc. just before the polarogram
information was gamed from the polarographic study. The diffusion current constant, I, was found to have a value of 2.4 + 0.1. This value of I is in agreement with that of 2.4 for lead tartrate complex (a presumably similar ion) in sodium hydroxide and thus indicates a 2 electron reduction_ That the reduction controlled was shown by the fact that ;/r/J E was constant to z%_ rial was-found to be -o_zg7 V ZIS. S.C.E_ and was independent
current was diffusion The half wave potenof the nickel and di-
_methylglyoxime concentrations over at least a ten fold rauge. Thus the reduction is probabl$ reversible and the same number of dimethylglyoxime ligands are associated with the oxidized (red) form and the mckel(IIj complex. A plot of log i/(&-i) ZJS.E yielded a straight line with a slope of 0.030, confirming a reversible z electron reduction_ On the basis of the above data the electrode reaction would appear to be hTi(IV)Dr
+
zz e Z+ Ni(Ii)D,
where x is 3 or more_ Actually no statement can be made ide-ion participates ‘m the electrode reaction. Variations
as to whether of hydroxide
/_ Elecivoanal.
Chem..
or not hydroxconcentration 8 (IgG4) 434-441
xICKEL(IV)
were
not
attempted
since
a large
DIMETHYLGLYOXIME
enough
change
437
to result
in detectable
half-wave
potential shifts could not be accomplished without making the red complex unstable. Although the polarographic data strongly indicates CoiatroZZed potential electrol~~sis a two electron reduction, further tests were desired to confirm the fact that a two electron reduction was indeed occurring and that the reduction was between Ni(IV) and Xi(II),
and did not involve
the reduction
of molecular
oxygen
bound
to the nickel
as suggested
in ref. 7_ with a mercury pool electrode, was run at X controlled potential electrolysis, -0-60 V us. S-C-E. This potential is reducing enough to reduce the red complex (see dimethylglyoxime or hydrogen peroxide Fig. I) but not so reducin g that nickel(I1) (which might result from the reduction of bound molecular oxygen) would react. Thesolutionscontained o.~~-~o-s.M of nickel(I1) and IO-IO-3Mof dimethylglyoximein IOO ml of 1.5 M NaOH. Complete osidation was accomplished by treatment with PbO-) for five hours_ The excess solid PbOz was removed by filtration just before the electrolysis_ Dissolved osygen was removed by a stream of nitrogen. The average gz value for the controlled potential reductions at the mercury cathode was 1-9. This value fell slightly short of 2.0 due to electrode fouling by nickel(I1) dimethylglyoxime precipitating near the end of the electrolysis and possibly to a decomposition of the red nickel(W) dimethylglyoxime complex. The solution resulting from the electrolysi- s was light yellow and had the same spectra as the nickel(I1) dimethylglyo_time solution made by mixing the air-free components. A “titanium spot test” for hydrogen pero-xide was negative_ By recording current-rel-ersal chronopotentiograms using a platinum working electrode it was found that air-free nickel(I1) dimethylglyoxime could be o_xidized V vs_ SCE. and a reverse (reduction) wave (having a transition time at E* = f-o.22 one-third that of the oxidation transition time) with an E+ = -0.34 V vs. S.C.E. could be recorded. The opposite case could also be accomplished, starting with nickel(IV)dimethylglyoxime. When the current was varied, the quantity z& was found to be constant. It may thus be concluded that the electrode reaction at platinum is diffusion controlled and of the simple irreversible typelo. On the basis of the above values of Ea, a controlled potential electrolysis study was initiated nickel(I1)
at a platinum electrode. An air-free solution containing O.O~OO-IO-~ M and 0.3-10-3 M of dimethylglyoxime in r.5 M sodium hydroxide was made
of up
under nitrogen in the controlled potential electrolysis cell. The passage of nitrogen was continued throughout the experiment. The solution was oxidized electrolytically at a controlled potential of to.40 V ‘us. S.C.E. The deep red color of the oxidized nickel dimethylglyoxirne complex rapidly appeared, indicating oxygen or oxygen producing species to be unnecessary_ The electrolysis current eventually dropped to a low but finite value (direct oxidation of dimethylglyoxime)_ After correction for this background current an 77 value of x.92 was calculated. The working electrode was now made the cathode, the controlled reduction potential being -0.40 V zs. S.C.E. As the electrolysis proceeded the red color faded to the original yellow. The IZ value for reduction was found to be r-99. No background correction was necessary. Since the oxidation and subsequent reduction of nickel in a solution of dimethylglyoxirne-r-5 M sodium hydrouide’ showed analytical promise for a range of nickel concentrations generally too high for spectrophotometric work, an accuracy and interference study was made. The electrolyte, the concentrations of which were 1.5 M J_ Ebclroanal.
Chem..
8 (1964)
434-441
ID. G.
435
DAVIS,
E.
A.
BOUDREAUX
in sodium hydroxide and 0.01 1M in dimethylglyo_xime, was placed in the cell. A nickel sample was added, some of which was oxidized by dissolved air. The oxidation was completed by controlled potential electrolysis at to.30 V z~rs. S C.E. The quantity of electricity passed was not recorded. TVhen the current dropped to a low constant value, the electrolysis was stopped and oxygen removed by bubbling nitrogen through the solution_ The oxidized nickel was then reduced at -0-40 V vs. S.C.E. and the coulombs used to accomplish this, recorded. Table I indicates the results obtained TABLE THE
COMTROLLED
POTESTIXL
1 DETERS!IS.~TIOX
OF
NICKEL
.-L&fed ion {trreqtriv.) ---
Nicket takem fmeqzris-)
A-zckelfozrnd fi??.eptV.)
O.OQ8Q
o.ogS3 o.ogS6
o-0500
a_ogos 0.0~00 o-0499 a-0499
Co’+
0.0622
bln”‘(o.02)
0.04s3
Fe=
o.ozSS
IF&+ (o.o=rr)
0.0296 0.0301
co’” x3*
O-0303
al’”
(o_a3) (0.03)
0_03az
ci-
(o.oc))
0_0300
(a_og) (o.02) (0.05)
and the effect of interferences. In general, results can be obtained with an accuraqof better than 0_5~~~_Manganese and iron appear to interfere, the latter apparently due to electrode fouling which causes incomplete o,xidation of the nickel. The controlled-potential electrolysis evidence seems to substantiate the idea that the red-colored material is, indeed, a nickel(IV) dimethylglyoxime complex, rather than one of nickel(.KII) or invoIving an oxygen carrying chelate of nickel(U). Some additional evidence was found through spectrophotometry and magnetic measure-
rixents.
of the main points in favor of the idea that the red colored complex was of the oxygen carrying chelate variety was that the color could be changed to light yellow by heating to about 70” or holding the solution under partial vacuum for a few minutes’_ At first thought, it would seem that these procedures remove oxygen from solution and thus reverse a reversible oxygen chelate formation_ -Indeed; if nitrogen is passed through a previousIy oxidized nickel dimethylglyoxime solution, the color also fades slowly (and the polarographic wave of Fig. I disappears). -In view of the controlled potential electrolysis experiments and the fact that dimethylglyoxime is oxidizable at a platinum electrode at potentials slightly more oxid&ing t&n necessary for the formation of nickel(W). it appears that the nickel(IV) dimetbylglyo~e complex is unstable. The nickel(IV) is a strong enough oxXizing agent to i3xidize dimethylglyoxime slowly, probably to _furoxane6. This would -&.c-explain the disappearance of the red color on heating, etc. Toir&stigate this reaction more carefully a sample of nickel(N) in excess oxygenSpi?Xt~OpJzOtOF?Z&-
One
J_ EZ~cfman~Z. Chem_. 8 frg64)
a$$-+#1
KICKEL(IV)
DIMETHYLGLYOXIME
439
free dimethylglyoxime solution, 1-5 M in sodium hydro.xide, was sealed in a spectrophotometer cell and spectra taken over a period of several days (Fig. 2). The final spectra is identical with that of unoxidized nickel(I1) dimethylglyoxime. The fact that the comples changes spontaneously to the yellow nickel(I.1) complex in a
I
320
t
360
I
400
I
440
I
400
Wovelength
I
5;o
560
I
600
.mp
t
640
6
0
spectra of 24.0- 10-5 ~11nickel in a solution 1-5 AT in Ku’aOH and .+- 10-1 .M in The solution was treated with oxygen. the e_xcess osygen removed with sealed in the absorption cell Curve A, taken immedlatel>-; curt-e 33. after Z+ h; curare C,aftcr 85 h (no further change observed.)
Fig. 2. Absorption dimethylgl~oxirne.
nitrogen
and
sealed vessel at room temperature indicates that the instability is only accelerated by heat rather than being caused by it. The application of vacuum simply removes dissolved oxygen which othenvise would continue to oxidize the nickel. This evidence again supports the idea that the red complex may well be a nickel(IV) species-the nickel(W) being a strong enough oxidizing agent to react with dimethy~glyoxime. Ox&Z AND &RNEK" have shown by means of photometric titrations that the red complex has a ratio of three dimethylglyoxime ions for each nickel ion, and by means of paper electrophoresis that the charge on the red complex is -2 _ Experiments in this laboratory tend to confirm these results. In the light of this work and the polarographic studies, the electrode reaction at mercury and probably platinum electrodes can be written
The nickel(IV)
complex
can then decompose
slowly at room temperature
where DO= is most likely furoxane. (An attempt to isolate furoxane was unsuccessful but it is a known oxidation product of dimethylglyoximell). Magnetic data In an effort to establish definitely that the oxidation state of nickel in the red complex was indeed +4, a magnetic spsceptibtity study was conducted. J_ Ekiroanal.
Chews.. 8 (1964)
434-441
440
D.
G.
D_4VIS,
E.
A.
BOGDREAGX
The results of these deterlminations yielded for the complex effective molar susceptibilities, X;?, of +640 . IO-~ and +-64s - 10-6 for nitrogen and oxygen saturated solutions respectively. (Solutions saturated in osygen were corrected for the susceptibility of the dissoLved oxygen) The magnetic moment calculated from these data is r.23 =f= o.oz B.M., assumin g a simple Curie Law behavior. The temperature variation within the range 2gg0studies showed a change in susceptibility of 2.4 -F_ 1.0% 33g”K, whereas the expected temperature change (assuming the Curie temperature relationship) for a paramagnetic ion with one unpaired electron should have been at least 4% within the limits of experimental error. Thus the observed temperature variation is just within the limits of experimental error, while that expected for one unpaired electron is a factor of three times greater than what could be attributed to experimental error. Hence we are forced to conclude on the basis of these limited studies that the susceptibility of Ni(IV)Da-2 is independent of temperature, at least within the restricted tenperature range of investigation and high dilution conditions_ Although the value determined for the magnetic moment at ambient temperature should be reasonably accurate, too much significance should not be attached, however, to the temperature variation studies If the susceptibility is indeed temperature independent than the susceptibility must be of the high frequency type, wherein the paramagnetism arises from mixing of a singlet ground state with low lying singlet excited states z&z orbital angular moments. However, the observed magneticmomentis too large by about a factor of two to account for this effect alone. Magnetic moments of the magnitude found here have also been reported for other with Ni(IV) compoundsls as well as for Ni(I1) complexes in planar four-coordination a diamagnetic ground state 13. In the latter cases, if the Ni(IT) ion is in a weak tetragonal field then the paramagnetism should arise from mixing of the singlet ground stat& with a thermally accessible triplet state of the complexl*. Although alternative mechanisms may be equally valid for explaining the anomalous magnetic behavior of four-coordinated Ni(IT) complexes 13, the thermal population of a tnplet level is the most reasonable explanation for the anomalous magnetic behavior of a six-coordinated Ni(IV) complex_ That ~e~y~glyo~e is a strong field @and is evidenced by the fact that Ni(II)Dz, an effectively- square planar complex, is diamagnetic. Since the effective ligand field stieng$h increases proportionally with increasing positive charge on the central metal atom, then octahedrally coordinated Ni(IV)Ds s- should certainly e,xist in a diamagnetic singlet ground state. Consequently, in addition to high frequency effects, the param &gnetism observed in this Iatter complex should result from the population of a thermally accessable triplet level. If this interpretation is correct then the magnetic moment should increase with temperature, which has not yet been compl+ely confirmed for the complex in solution. Future investigations will include the attempt to isolate the complex and subject it to criti&I- magnetic and spectral investigations, Theoretical correlation with the experimentalobservables will also be attempted in an effort to gain a better under~standing of the bonding- ih this complex_ ACKNOWLEDGEMENTS
iavesrigation from the N&bnal
This
We sr.Qported in part by Public H&a&h Service Grant AM08z48 Instituttit of Arthritis and_ Metabolic D_iseases. J_
EZtztvounaL
Cliem.,‘El
(Ig6q)
43=9-+4x
s;ICKEL(IV)
44=
DIMETHYLGLYOXIME SUMMARY
The
red solutions
of nickel
dimethylglyoxime
in strongly
basic
solution
have
been
studied by electrochemical, spectrophotometric and magnetic methods. It is concluded that the red complex is a species of Ni(IV). A controlled potential coulometric method for the analysis of nickel is also presented. REFERENCES I F. FEIGL, Ber.. 57 (1924) 75S_ 2 G HAIM AID l3. TXRRANT, Ind. Etzg. Clrertr., IS (lg@) 51. AcLa Chettr. Scnnd., 3 (1949) 497. @I. 3 I<_ Ax JASSEN AND EL XYGA_XRD, 4 hl. HOORE>IAN, _4maZ. Chim. Ada, 3 (1949) 635. =j A. OK&? AND &I. SIR-VEK, CO~~C~~O?Z CxCh. Chel?a.Comm~m_, 24 (1959) 1-699. Ed B. JEZOWSK_~-TRZEBI_~TOWSK_~, 6 -4. OI-&Z, Theovy alrd StrmLzcve oj Complex Compourz~, Pergamon Press Inc.. New York. 1964. p- 167~ 7 J_ SELBINAKD J_ H_ Jus~~rv.J..~m. CAenz. SOL. Sz (rgGo) 1057_ 8 D. G. D_4vrs, J_ EZecrroaxaZ. Clretn., I (1959) 73_ g D G. DAVIS. =Inal. Chem.. 33 (IgGI) 1839. IO W. REIN~IUTH. rival. C7lenz.,32 (IgGo) rgr+ Ed. L. MEITLS. BlcGraw-Hill Book Co. II I?. C. &I_4RKEx_4S, Hat&book of AtzaZyticaZ Chettzislry. Inc., I\l;ewYork, 1963, section I?-,p- 14'. I?- I?.ROY _4x~ B. S_~R~I,I~, J_ I?zdiau Chenr. SOL, z=j (rg+S) zo=j_ 13 S. L.HoLT, R. J. BOUCHARD ASD R-L. CXRLI~, J. Am.Chenr. Sot., SG (IgGa) 519. 14 C. J_ B_~LLH=~USE~ .4i‘;~ X D_ LIEHR. J_ Am. Chetn. Sot., SI (1939) 535. J_ Eleclroatzal.
Chettz.,
8 (IgGq)
434-441