Thermal decomposition of potassium hexanitronickelate(II) hydrate

Article abstract of DOI:10.1016/j.tca.2005.02.018

The thermal decomposition of the complex K₄[Ni(NO ₂)₆]·H₂O has been investigated over the temperature range 25-600 °C by a combination of infrared spectroscopy, powder X-ray diffraction, FAB-mass spectrometry and elemental analysis. The first stage of reaction is loss of water and isomerisation of one of the coordinated nitro groups to form the complex K₄[Ni(NO ₂)₄(ONO)]·NO₂. At temperatures around 200 °C the remaining nitro groups within the complex isomerise to the chelating nitrite form and this process acts as a precursor to the loss of NO₂ gas at temperatures above 270 °C. The product, which is stable up to 600 °C, is the complex K₄[Ni(ONO) ₄]·NO₂, where the nickel atom is formally in the +1 oxidation state.

Full text of DOI:10.1016/j.tca.2005.02.018

Thermochimica Acta 429 (2005) 227–231
Thermal decomposition of potassium
hexanitronickelate(II) hydrate
Elif Subasia,∗, Matthew J. Almondb
a
Dokuz Eylu¨l University, Faculty of Science and Arts, Department of Chemistry,
˙
Buca Tinaztepe, 35160 Izmir, Turkey
b
Reading University, Department of Chemistry, Whiteknights, Reading RG6 6AD, UK
Received 12 January 2005; received in revised form 9 February 2005; accepted 14 February 2005
Abstract
The thermal decomposition of the complex K4[Ni(NO2)6]·H2O has been investigated over the temperature range 25–600 ◦C by a combination
of infrared spectroscopy, powder X-ray diffraction, FAB-mass spectrometry and elemental analysis. The first stage of reaction is loss of water
and isomerisation of one of the coordinated nitro groups to form the complex K4[Ni(NO2)4(ONO)]·NO2. At temperatures around 200 ◦C the
remaining nitro groups within the complex isomerise to the chelating nitrite form and this process acts as a precursor to the loss of NO2 gas at
temperatures above 270 ◦C. The product, which is stable up to 600 ◦C, is the complex K4[Ni(ONO)4]·NO2, where the nickel atom is formally
in the +1 oxidation state.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Thermal analysis; Nickel(II); Nitro; Potassium salt; X-ray diffraction; IR data
1. Introduction
While most hydrated compounds can be dehydrated without
inducing other changes, it was confirmed that the dehydra-
tion of orange K4[Ni(NO2)6]·H2O at 100 ◦C produces the
red complex K4[Ni(NO2)4(O2N)]·NO2 [13]. According to
the published data rehydration of the hygroscopic mixture at
≈20 ◦C causes K4[Ni(NO2)6]·H2O to be reformed [13]. The
nickel(II) in K4[Ni(NO2)4(O2N)]·NO2 has a distorted octa-
hedral ligand environment, with four nitrites co-ordinated via
nitrogen, and a fifth chelating via the two oxygen atoms [13].
Although dehydration and linkage isomerisation have been
studied in detail, a complete thermal decomposition analy-
sis of the complex has never been explored. In the present
work the thermal dehydration, isomerisation and decompo-
sition behaviour of potassium hexanitronickelate(II) hydrate
(K4[Ni(NO2)6]·H2O) over the temperature range 25–800 ◦C
has been investigated. A further decomposition reaction has
been discovered in this work, which occurs between 235 and
275 ◦C and the nature of this reaction is discussed in terms
of thermal data, elemental analysis, atomic absorption spec-
troscopy, infrared spectroscopy, X-ray diffraction and mass
spectrometry.
Because of its capability of bonding to metal atoms in a
variety of different ways, the nitrite ion in complexes has
been of interest for many years and a range of nickel(II)
polynitritecomplexesofdifferingcompositionhavebeenpre-
pared [1–6]. It has been known for 150 years that addition
of nickel(II) ions to an aqueous solution of potassium nitrite
produces an orange crystalline precipitate [7]. This product
was originally formulated as K4[Ni(NO2)6] although it was
subsequently shown that the orange compound is in fact a
monohydrate. The compound K4[Ni(NO2)6]·H2O contains
the [Ni(NO2)6]4− ion with 6N bonded nitro ligands. IR and
electronic spectra suggest that the water molecule is not di-
rectly bonded to the nickel centre [8,9]. The dehydration of
K4[Ni(NO2)6]·H2O and the simultaneous linkage isomerisa-
tion has been studied using TGA and DSC [10,11] and the ki-
netics of the isomerisation process has been investigated [12].
∗
Corresponding author. Tel.: +90 232 453 50 73; fax: +90 232 453 41 88.
E-mail address: elif.subasi@deu.edu.tr (E. Subasi).
0040-6031/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2005.02.018

228
E. Subasi, M.J. Almond / Thermochimica Acta 429 (2005) 227–231
2. Experimental
2.1. Instrumentation
2.1.1. Thermal analysis unit
A Stanton Redcroft thermal analysis system (TG/DTA-
32), with a temperature-programmable thermal balance, and
a platinum crucible was used for obtaining TGA and DSC
thermograms in air. The rate of heating was fixed at 10 K/min.
The sensitivity of the instrument is 0.1 mg.
2.1.2. Infrared spectra
IR spectra of solid complexes in KBr discs were recorded
using a Mattson 1000 FT interferometer. A mixture of ap-
proximately 5% of the sample with KBr was ground in a
pestle and mortar and pressed into a disc. Typically 16 scans
were run at 4 cm−1 resolution for each spectrum.
Fig. 1. TG/DTA curve of potassium hexanitronickelate(II) hydrate.
ing at an accelerating voltage of 10 kV. Samples were des-
orbed from a nitrobenzyl alcohol matrix using 3 keV xenon
atoms.
2.1.3. X-ray diffraction data
X-ray powder diffraction (XRD) patterns were recorded
˚
at the Cu K␣1 wavelength of 1.540598 A on a spectrolab CPS
2.1.6. The apparatus for the collection of gaseous
products on the vacuum line
120 Series 3000 X-ray diffractometer attached to a micropro-
cessor equipped with an Inel Multichannel position-sensitive
detector. Samples were ground into a very fine powder before
analysis.
A glass vacuum line fitted with Young’s taps was used for
the thermal decomposition experiment, in order to sample
gaseous products. In this apparatus a speedivac mercury dif-
fusion pump, backed by a high vacuum rotary pump, enabled
a pressure of ca. 1 × 10−5 Torr to be attained. A 0.27 g sam-
ple of K4[Ni(NO2)6]·H2O was placed in an evacuated reactor
(volume 200 cm3) fitted with a break seal. The reactor was
removed from the vacuum line after sealing and was left in an
oven at 280 ◦C overnight. After this time the reactor was left
to cool before being fitted with a tap, joint and side arm con-
taining a ball bearing to open the break seal. This apparatus
was transferred back onto the vacuum line and the connector
was evacuated before the seal was broken. The gaseous prod-
ucts were transferred to a gas cell fitted with KBr windows
for IR spectroscopic analysis.
2.1.4. Elemental analysis
CHN analyses were performed at The Technical and Sci-
entific Research Council of Turkey, TUBITAK by burning the
sample at high temperature (1000–1100 ◦C) and measuring
CO2, H2O and N2 concentrations on a LECOCHNS-0-9320
instrument. Potassium and nickel concentration were mea-
sured at The University of Reading using a Sherwood Model
410 Flame Photometer and a Perkin-Elmer Model 1100B
Atomic Absorption Spectrophotometer, respectively.
2.1.5. FAB-mass spectra
Fast atom bombardment (FAB) mass spectrometry is a
technique in which samples are ionised by bombardment
with a beam of highly energetic atoms of (for example)
xenon or argon. Samples are typically dissolved in a non-
volatile solvent in order to reduce the lattice enthalpy that
must be overcome if the ions are to be liberated from a
sample. FAB-ionisation techniques are particularly suitable
for analysing higher molecular weight samples. In the cur-
rent work positive-ion FAB-mass spectra were recorded at
TUBITAK on a JEOL SX102 mass spectrometer operat-
2.2. Preparation and analysis
The K4[Ni(NO2)4]·H2O complex salt was prepared by
adopting the method of Goodgame and Hitchman [9]. A con-
centrated aqueous solution of Ni(NO3)2·6H2O was added to
a saturated solution of KNO2. The latter compound was in
excess. The solid product was separated by filtration and was
dried for several days in a desiccator.
Table 1
Summary of the thermal decomposition of the potassium hexanitronickelate(II) hydrate
Compound
ꢀTmin (◦C)
Step number
Temperature (◦C)
Loss in weight (%)
Starting
Ending
103.0
Observed
3.40
Calculated
3.54
K4[Ni(NO2)6]·H2O
100
220
270
1
2
95.2
235.7
273.4
9.00
9.04

E. Subasi, M.J. Almond / Thermochimica Acta 429 (2005) 227–231
229
3. Results and discussion
been used to identify the manner of nitrite coordination in a
complex, because the frequencies of the bonds change from
the free ion values depending upon the type of nitrite co-
ordination such as nitro, nitrito, chelating (asymmetrically
or symmetrically) nitrito and bridging nitrite [14–16]. IR
spectra and XRD traces show clearly that there is a distinct,
hitherto unreported compound which is formed at around
270 ◦C from the more well-known linkage isomerism prod-
uct K4[Ni(NO2)4(ONO)]·NO2 which is produced at lower
temperatures. The infrared data are summarised in Table 2.
In the FT-IR spectra of the compound before and after
heating to 100 ◦C are in a good agreement with those re-
ported previously [9] for the complexes K4[Ni(NO2)6]·H2O
andK4[Ni(NO2)4(ONO)]·NO2, respectively. Thedehydrated
complex shows peaks with similar positions and relative in-
tensities to those in the monohydrate, at 810, 826, 1321 and
1346 cm−1 and these are assigned to the nitro groups in the
[Ni(NO2)4(O2N)]3− complex anion. In addition, a peak of
moderate intensity occurs at 1270 cm−1 assigned to νasym
3.1. Thermal analysis
The TGA and DTA traces and summary of the thermal de-
composition of potassium hexanitronickelate(II) hydrate are
shown in Fig. 1 and Table 1, respectively. The curves show
the endothermic loss of coordinated water around 100 ◦C ac-
counting for 3.40% the mass of the sample. The calculated
mass loss for one molecule of water from this complex is
3.54% but this discrepancy is not thought to be significant in
the light of previous experiments [10–12]. In one study, dehy-
dration of a large number of samples of K4[Ni(NO2)6]·H2O
gave an average mass loss of 2.68% while in other work
[10,11] mass loses of 3.3 or 3.9% were obtained. It appears
that the actual H2O content depends somewhat on preparation
and drying conditions.
The anhydrous product is stable up to about 200 ◦C. A sec-
ond endothermic change, with no mass loss, occurs at around
220 ◦C. This process probably corresponds to the isomerisa-
tion of the remaining nitro groups. At around 270 ◦C a third
endotherm is accompanied by a mass loss of ca. 9.00%. It ap-
pears likely that this process corresponds to the loss of NO2
gas; the calculated mass loss for one NO2 molecule is 9.04%.
Upon further heating the mass of the sample remains constant
up to 600 ◦C.
In an attempt to verify that linkage isomerism does indeed
occur at around 220 ◦C and to characterise the product of
the thermal reaction occurring around 270 ◦C, a sample of
K4[Ni(NO2)6]·H2O was heated in an oven. After heating to
temperatures between 200 and 500 ◦C for several hours in a
step by step manner, samples were removed and were ana-
lyzed by infrared spectroscopy and powder X-ray diffraction.
−
of NO2 ion which is not coordinated to the nickel centre.
The IR spectrum of KNO2 was obtained for comparison and
shows νasym as an intense broad peak at 1260 cm−1. The re-
maining peaks seen for the dehydrated product at 866 cm−1
,
and shoulders at ≈1225 and 1385 cm−1 are attributed to
the chelating nitrito group in the [Ni(NO2)4(O2N)]3− an-
ion. This assignment is confirmed by examination of an
earlier publication [9] which reports the spectrum of pure
K3[Ni(NO2)4(O2N)] prepared by extraction of KNO2 from
the anhydrous residue. Here the peak at 1271 cm−1 assigned
−
to NO2 almost disappears, and the peaks at 1225 and
1385 cm−1 assigned to νasym and νsym of the chelating nitrite
are much more clearly resolved, demonstrating that overlap
did indeed occur in these spectral regions for the mixed sam-
ple.
3.2. Infrared spectra and powder X-ray diffraction
traces of potassium hexanitronickelate(II) hydrate and
the decomposition products
Upon heating to temperatures around 200 ◦C infrared
bands arising from monodentate N-bonded nitro groups de-
cay to extinction while those arising from chelating ni-
troso groups increase in intensity. At this temperature a
second endotherm, with no corresponding loss of mass is
The infrared spectrum of the nitrite ion, in particular the
symmetric and asymmetric stretching frequencies, has often
Table 2
n−
Nitrite infrared frequencies (cm−1) for the thermal decomposition products of potassium hexanitronickelate(II) hydrate and some M(NO2)x anions
Compound
νs(NO2)
νas(NO2)
νs(ONO)
νas(ONO)
δ(NO2)
δ(ONO)
K4[Ni(NO2)6]·H2O
1346 (m)
1346 (m)
1319 (s)
1321 (s)
–
–
831 (s), 810 (w)
826 (s), 810 (w)
826 (w), 810 (w)
826 (s), 810 (m)
826 (vs), 810 (w)
810 (s)
Heated at 100 ◦C
1385 (sh)
1395 (s)
1395 (s)
1395 (vs)
1225 (sh)
1265 (s)
1265 (s)
1265 (vw)
1265 (b)
1265 (b)
866 (s)
Heated at 200–230 ◦C
Heated at 270–405 ◦C
Heated at 410–520 ◦C
KNO2
KNO2 heated at 270 ◦C
810 (s)
749
a
NO2(g)
1610
1318
b
Cs2Zn(NO2)4
1381 (s)
c
1167 (vs)
1258 (vs, b)
1266 (vs, b)
847 (m)
858 (w), 851 (m)
858 (w), 851 (m)
b
K3[Hg(NO2)4]·NO3
b
c
K2Cd(NO2)4
b: broad; m: medium; s: strong; sp: sharp; sh: shoulder; w: weak; vs: very strong; vw: very weak.
a
All data were obtained in inert gas matrices.
Taken from Ref. [17].
b
c
Band obscured by νas
.

230
E. Subasi, M.J. Almond / Thermochimica Acta 429 (2005) 227–231
Table 3
Values from mass spectra of thermal decomposition product
Complex
Thermal decomposition product formed at 270 ◦C
MW
445
Relative intensities of ions and
assignments
430 (5) [M+ − O]; 384 (7) [M+ − O (NO2)]; 345 (44) [M+ − O (NO2) K]; 276
(17) [M+ − O (NO2) 2K NO]; 230 (45) [M+ − O 2(NO2) 2K NO]; 191 (100)
[M+ − O 2(NO2) 3K NO]; 115 (9) [M+ − O 3(NO2) 3K 2NO]
For the mass spectral data relative intensities are given in parentheses; probable assignments in square brackets. M+ refers to the parent molecular ion
{K4[Ni(O2N)4]·NO2}+ and the assignments show the loss of specific atoms and groups from this ion.
Table 4
Elemental analysis results (%)
Complex
Found (calculated) (%)
K
Ni
N
H
K4[Ni(NO2)6]·H2O
30.53 (30.65)
34.93 (35.06)
11.45 (11.59)
13.11 (13.26)
16.44(16.50)
15.62 (15.73)
0.36 (0.39)
–
Thermal decomposition product at 270 ◦C
seen. These findings indicate that isomerisation of the re-
maining nitro groups within the complex occurs, forming
K4[Ni(O2N)5]·NO2 which contains the [Ni(O2N)5]3− ion.
Thereareprecedentsforaformulationofacomplexionwhere
all of the NO2 groups adopt a chelating bidentate structure,
e.g. in the mercury(II) complex K3[Hg(O2N)4]·NO3 and the
zinc(II) complex Cs2[Zn(O2N)4] [17].
4. Conclusion
On the basis of the studies reported here it is possible
to propose a mechanism for the thermal decomposition of
K4[Ni(NO2)6]·H2O over the temperature range 25–600 ◦C.
Around 100 ◦C dehydration occurs to give rise to the complex
K4[Ni(NO2)4(O2N)]·NO2. Above 200 ◦C a further isomeri-
sation takes place such that all of the groups adopt a chelating
bidentate structure. This step appears to be a crucial precur-
sor to the subsequent loss of NO2 gas above 270 ◦C and the
concomitant reduction of the nickel(II) centre to nickel(I).
Upon heating to temperatures above 270 ◦C, the IR fea-
tures assigned to the complex K4[Ni(O2N)5]·NO2 decay. A
gaseous product is also formed which was studied by IR spec-
troscopy. The spectrum of this gas shows peaks at around
1600 and 1300 cm−1. These identify the product as NO2 [14].
The next stage of the investigation was to attempt to
identity the solid product formed by loss of NO2 gas. In-
frared spectra of the solid product show the presence of
chelating ONO groups, as evidenced by bands at 1395 cm−1
(νs(ONO)), 1265 cm−1 (νas(ONO)) and 826 and 810 cm−1
(δ(ONO)) [14–17]. There is no evidence for any monoden-
tate nitro (NO2) groups. It therefore seems likely that all of the
NO2 groups in this product adopt a chelating nitro coordina-
tion. The product shows the following features in the powder
Acknowledgements
We thank the Turkish Republic Ministry of Education for
a studentship for Elif Subasi at University of Reading. We
are grateful to Professor D.A. Rice for helpful discussions
and Mr. A. Jahans for carrying out the X-ray analysis of the
powder samples.
˚
X-raydiffractionpattern(d-spacingsmeasuredinAwithrela-
tive intensities in parentheses): 4.922 (76), 4.035 (100), 3.183
(32), 2.975 (57), 2.711 (35), 2.562 (30), 2.365 (33), 2.307
(62), 2.251 (21), 2.101 (32), 2.012 (14), 1.822 (9), 1.495 (7),
1.237 (6) and 1.198 (14). This pattern does not correspond to
that of the high-temperature cubic form of KNO2 [18a] nor
to that of any other known form of KNO2 [18b–d]. Thus it is
shown that the product is not a mixture containing KNO2, but
rather a single phase best formulated as K4[Ni(O2N)4]·NO2.
ThisobservationisinkeepingwiththeFAB-massspectrumof
theproduct(Table3)whichshowsfragmentationoftheparent
molecular ion {K4[Ni(O2N)4]·NO2}+ and with the results of
elemental analysis (Table 4). This means that the nickel atom
is in the unusual but not unknown (e.g. [Ni(PMe3)4][BPh4]
[19]) oxidation state of +1. The disproportionation reac-
tion in which the nickel centre is reduced from the +2 to
the +1 oxidation state is driven by the production of NO2
gas.
References
[1] M.A. Hitchman, G.L. Rowbottom, Coord. Chem. Rev. 42 (1982) 55.
[2] D.M.L. Goodgame, M.A. Hitchman, Inorg. Chem. 3 (1964) 1389.
[3] A.J. Finney, M.A. Hitchman, C.L. Raston, G.L. Rowbottom, A.H.
White, Aust. J. Chem. 34 (1981) 2085.
[4] S. Tagaki, M. Joesten, P. Galen-Lenhert, Acta Crystallogr., Sect. B
31 (1975) 1970.
[5] I. Nakagawa, T. Shimanouchi, K. Yamasaki, Inorg. Chem. 3 (1964)
772;
I. Nakagawa, T. Shimanouchi, K. Yamasaki, Inorg. Chem. 7 (1968)
1332.
[6] B.J. Hathaway, R.C. Slade, J. Chem. Soc. A (1966) 1485.
[7] N.W. Fisher, Poggendorffs Annalen 74 (1848) 115.

E. Subasi, M.J. Almond / Thermochimica Acta 429 (2005) 227–231
231
[8] A. Ferrari, L. Cavalca, M. Nardelli, Gazz. Chim. Ital. 81 (1951) 945;
M. Le Postollec, J.P. Mathieu, Comp. Rend. 254 (1962) 1800;
M. Le Postollec, J.P. Mathieu, H. Poulet, J. Chim. Phys. 60 (1963)
1319.
[9] D.M.L. Goodgame, M.A. Hitchman, Inorg. Chem. 6 (1967) 813.
[10] J.E. House, R.K. Bunting, Thermochim. Acta 11 (1975) 357.
[11] J.E. House, Thermochim. Acta 32 (1979) 139.
[12] J.E. House, J.K. Muchling, C.C. Williams, Thermochim. Acta 53
(1993) 222.
[13] E. Grey, M.A. Hitchman, G.L. Rowbottom, N.V.Y. Scarlett, J. Wil-
son, J. Chem. Soc., Dalton Trans. I (1994) 595.
[14] K. Nakamoto, IR and Raman Spectra of Inorganic and Coordination
Compounds, 1986, p. 309.
[15] B.J. Hathaway, R.C. Slade, J. Chem. Soc. A (1966) 485.
[16] D.M.L. Goodgame, M.A. Hitchman, D.F. Marsham, J. Chem. Soc.
A (1970) 1933.
[17] D.M.L. Goodgame, M.A. Hitchman, J. Chem. Soc.
612.
[18] (a) JCPDS file No. 27-429;
(b) JCPDS file No. 22-840;
(c) JCPDS file No. 26-1326;
(d) JCPDS file No. 27-428;
(e) JCPDS file No. 27-430.
A (1967)
[19] A. Gleizes, M. Dartiguenave, Y. Dartiguenave, J. Galy, H.F. Klein,
J. Am. Chem. Soc. 99 (1977) 5187.