Studies on the formation and thermal properties of manganese phosphate coatings and hureaulite by means of conversion electron and transmission M6ssbauer spectrometry KIYOSHI NOMURA, YUSUKE UJIHIRA
Faculty of Engineering, The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo, 113 Japan
The chemical states of iron species, produced in manganese phosphate coatings by the immersion of steel in a manganese phosphate bath, were analysed with the help of conversion electron and conventional transmission M6ssbauer spectrometry (CEMS and TMS). The modifications of hureaulite by the variation of iron to manganese ratio are discussed with regard to the growth of manganese phosphate coatings. The thermal properties of manganese phosphate coatings and synthetic hureaulite were also investigated by means of differential thermal analysis and the composition analysis of thermal products.
1. Introduction The phosphating of metals and alloys is now widely used as a means to obtain a corrosionresistant surface, which also serves as an appropriate surface finishing for lubrication and painting. Various types of phosphating - zinc, manganese, iron and calcium phosphatings, and their modified phosphatings- are in current use . Coatings, which are deposited on a steel surface by the zinc, manganese, calcium, and calcium-zinc phosphatings, consist mainly of hopeite and phosphophyllite, hureaulite, brushite and monetite, and scholzite, respectively , whereas the main constituent produced by iron phosphating is amorphous paramagnetic iron ( I I I ) c o m p o u n d . Hitherto, the chemical state and epitaxial growth of iron compounds produced at the steel surface by the zinc and calcium-zinc phosphatings, the interfacial state of the substrate [4, 5], the effect of modest heating of phosphate coatings , and the thermal deterioration of zinc phosphate coatings  have been investigated by the 0022-2461/82] 123437-08503.46/0
authors and Berry by means of conversion electron M6ssbauer spectrometry (CEMS). The thermal transformation and the decomposition of synthetic phosphophyllite , and the chemical state of iron in synthetic iron (II) substituted hureaulite  were also studied using conventional transmission M6ssbauer spectrometry (TMS). In this paper, we describe the results of the characterization of phosphate coatings obtained by the manganese phosphating. The thermal properties of the coatings and the pyrolysis products of hureaulite are also investigated. The range of low-energy electrons is estimated by the application of Fetdman's equation  to hureaulite, MnsH2(PO4)4" 4H20. R (A) = 0.077 x E (keV)4"6. Therefore, the ranges of 7.3 keV K-, 13.6 keV L-conversion and 5.5keV Auger electrons reemitted by the internal conversion process of an S7Fe nucleus after the resonant absorption of 14.4 keV 7-rays are 720, 12600 and 200A (72,
9 1982 Chapman and Hall Ltd.
TAB L E I Comparison of M6ssbauer parameters for a scraped sample (78 K) from a manganese phosphate coating, calculated by a computer program with and without the restriction of lorentzian parameters. The spectrum was assumed to consist of three kinds of doublet peaks Line no.
2.1. Manganesephosphating procedure and preparation of hureaulite Rimmed steel sheets (SPCC-D) were degreased, washed, activated in the bath of Parcolene Z (Nihon Parkerizing Co Ltd), and immersed in a manganese phosphate solution at 95 ~ C for several minutes. After the phosphating, the sheets were washed with water and dried by air-blowing at about 110 ~ C for 2 min. The average thickness of the coatings was estimated from the weight difference of the phosphated steels before and after stripping off the coated layers by immersing them in a solution of 5% chromium (VI) oxide at 70~ for 30 min. The contents of manganese and iron in the phosphate coatings were determined by atomic absorption spectrometry. Synthetic hureaulite compounds were prepared b y adding lOOml of l mol/litre manganese sulphate and iron (II) sulphate solutions in 400 ml o f 0.3mol/litre sodium mono-hydrogen phosphate solution at a temperature higher than 9 5 ~ in a stream of nitrogen to prevent oxidation of iron (II) [91.
0.31 0.39 0.42 0.33 0.42 0.34
-+0.12 -+0.54 -+0.82 -+-0.08 +- 0.16 +- 0.22
1260 and 20 nm), respectively. Because electrons are also emitted from surface materials through the photo-electric and the Compton effects b y the irradiation o f the resonant scattered X- and 7-rays, the analytical information on surface layers of micron thickness can be obtained b y CEMS. The variation o f the composition of manganese phosphating layers is also discussed in relation to the phosphating period.
FWHM (mm sec -1)
2.2. Heat treatment of hureaulite Thermal dehydration properties of the synthetic hureaulite and the phosphate compounds deposited at the manganese phosphated steel were studied using Rigaku simultaneous D T A / T G A thermoanalyser, analytical reagent grade alumina being used as a weight standard. The thermal behaviour o f each sample was measured from room temperature to about 1000 ~ C, raising the temperature at a rate of 10 ~ C rain -1 . The temperatures quoted were determined by the maximum values from in the DTA curve. 2.3. Measurement of M/Jssbauer spectra Conversion electron M6ssbauer ( C E M ) s p e c t r a o f manganese phosphated steel surfaces were observed by detecting resonantly re-emitted electrons with a back-scatter-type gas-flow counter, coupled to a M6ssbauer spectrometer. A 30 mCi STCo (Rh) source was used. Transmission M6ssbauer (TM) spectra of the powder sample, prepared by scraping off the phosphate coating from the steel, and the heated hureaulite samples were measured by detecting transmitted "y-rays by an NaI(T1) scintilation counter. A 1 0 m C i STCo (Rh) was used. The velocity was calibrated by measuring a SVFe enriched a-iron foil. TM spectra were analysed by a computer program  with the restriction that all peaks have uniform FWHM and that the doublet peaks have equal intensity. The results of the calculation with and without the restriction are given in Table 1 for comparison. The former gives the more reliable M6ssbauer parameters.
"""'" "" .... "
[ Fe (III M(2)
M(1) ..";: M (3)
9 :.... - .... ., ..-." ..'.' ,-. ........ .
Figure 1 M6ssbauer spectra of a powder sample scraped from a manganese phosphated steel surface. Measuring temperature: (a) Room temperature; (b) liquid N 2 temperature.
(a) Room "". temperature :,
=: 89 o
. ...-=. 9
(b) 78 K
, i .9 9
0 2 Velocity ( mm sec-I )
The iron (Ill) species is assumed to result from the iron (II) ion which has been released from the steel and further oxidized by the nitrate ion present in the phosphate bath. The M6ssbauer parameters of iron species in manganese phosphate coating are calculated and summarized in Table II.
3.1. Chemical state of iron in manganese phosphate coatings The TM spectra of the scraped sample from manganese phosphate coatings, measured at room and liquid nitrogen temperatures, are shown in Fig. 1. The M6ssbauer spectra are composed of three doublets showing the presence of three kinds of paramagnetic iron (II) species, the M6ssbauer parameters of which agree well with those of iron (I1)-substituted hureaulite  at three sites of octahedral crystal structure [12, 13], and a doublet of paramagnetic iron (III) species, which is considered to be trapped in the coatings during the formation of manganese phosphate layers.
3.2. Change of composition of manganese phosphates coatings with immersion time in the phosphating bath The variation of weight and the contents of manganese and iron in the manganese phosphate coatings with immersion time (0.5 to 20 min) are shown in Figs 2 and 3, respectively9
TAB LE I I M6ssbauer parameters of powder samples scraped from manganese phosphate coatings Temperature
I.S. (mm sec- ~)
Q.S. (mm sec- 1)
1.244 1.23 1.24 0.3
• • • •
0.005 0.03 0.05 0.3
2.487 1.49 1.24 0.5
• 0.007 • 0.05 _+0.09 • 0.6
1.382 1.36 1.35 0.49
• • • •
0.005 0.01 0.01 0.01
2.81 1.98 1.64 0.71
• • • •
0.008 0.02 0.02 0.02
FWHM 0.323 + 0.005
State of iron M(2) Fe (II)* M(1) M(3) Fe (III) M(2) Fe (II)* M(1) M(3) Fe (III)
*Iron (II) states correspond to iron (II) ions substituted at three manganese sites [M(1), M(2) and M(3)] of octahedrons in hureaulite structure . 3439
'E o 5
E ~4 ..c
~ 3 ol r
Phosphating time ( r a i n )
Figure 2 Relationship between phosphating time and weight of manganese phosphate coatings.
At the initial stage of formation of manganese phosphate layers, the weight of the coatings and the iron content in the coatings increase linearly with the duration of phosphating. After 10 min immersion, the greater part of the coatings are deposited. The coating thickness is about 5 mg cm -2, which corresponds to 15.4/~m hureaulite p = 3.24), The weight of coating remains almost constant with the progress of manganese phosphating. However the manganese content in the coating increases, whereas the iron content decreases. The sum of manganese and iron contents in the coatings is almost 34% irrespective of the duration of the phosphating. The figure is less than the stoichiometric value of manganese and iron contents in hureaulite (36%). All X-ray diffraction patterns of the man-
m 20 r OJ
ganese phosphate coatings indicate the presence of hureaulite as a major constituent. The coatings are considered to be composed of a large amount of hureaulite with various fractions of iron and manganese in addition to a small amount of the other phosphate compounds. The CEM spectra of the manganese phosphate coatings are shown in Fig. 4. In the CEM spectra of the samples phosphated for 2.5 and 5 min, the sextet peaks of the substrate steel and the doublet peaks of paramagnetic iron (III) compound are seen. The doublet peaks of paramagnetic iron (II) species appear in addition to the peaks of the iron (III) species and substrate for the samples phosphated for 10 and 20min. Three doublet peaks of paramagnetic iron (II) species are assigned to the iron ions substituted hureaulite reported previously . As shown in Fig. 4c and d, the relative peak intensity of the two inner doublets [M(1) + M(3)] of hureaulite formed on the steel by 20min phosphating is higher than that of hureaulite formed by 10 rain phosphating. A previous study revealed that the highness of the relative peak intensity of inner doublets of hureaulite is attributed to the fact that iron (I|) ions preferentially occupy the M(1) and M(3) sites of the octahedrons of the hureaulite crystal lattice . The results of CEMS study indicate that the main composition of manganese phosphate coating changed from hureaulite with high iron (II) content to hureaulite with low iron (II) content after the saturation of coating, This conclusion is also confirmed by the determination of manganese and iron contents in the manganese phosphate coating. In the course of formation of the phosphate layer, iron-enriched hureaulite begins to grow, and after the steel surface is almost perfectly covered with a leading phosphate compound such as hureaulite, iron ions in hureaulite begin to be substituted by manganese ions in the phosphate solution. The hureaulite with low iron (II) content is formed last as a manganese phosphate coating.
3.3. Estimation of coating thickness detected by CEMS t
B 12 16 20 Phosphating time ( rain )
Figure 3 Relationship between phosphating time and contents of manganese and iron in manganese phosphate coatings.
Tile spectrum of iron located beneath a manganese phosphate coating of about 16/Jm thickness, which is thicker than the range of L internal conversion electrons in hureaulite, is observed. The phenomenon is due to polycrystallinity, porousness and non-uniformity of the manganese phosphate
Figure 4 Variation of conversion electron
2 . 5 m in
M6ssbauer spectra of manganese phosphate coatings with duration of phosphating. (a) 2.5 rain, (b) 5 rain, (c) 10 rain, (d) 20 min.
[d ) 20min .s
....... ".:.":'"~'?::...:~'.::. 9
Velocity ( mm sec -1 ) coating (Fig. 5). Conversion and Auger electrons re-emitted at the interface of steel and coating can pass through the vacant space and come out into the counter, in addition to the contribution of secondary electrons emitted by the irradiation of scattered 3'- and X-rays to the surrounding materials9
Figure 5 Scanning electron microscopic image of manganese phosphate coating.
The degree of fixation of the coated substances to the steel surface is estimated through the intensity ratio among the magnetically split peaks of substrate iron. Since the ratio of six peaks of iron was 3 : 2.5 : 1 : 2.5 : 3 when the steel was phosphated for 10rain, whereas the theoretical ratios is 3 : 4 : 1 : 1 : 4 : 3 when the magnetic orientation of surface iron is perpendicular to incident 7-ray . The phosphate compounds deposited at the steel surface was considered to be tightly fixed to the steel surface. 3.4. T h e r m a l t r a n s f o r m a t i o n o f h u r e a u l i t e The manganese phosphate coating, which contained 7% iron and 27% manganese, was mechanically scraped off from the phosphated steel surface and subjected to the thermal analysis by DTA. The result is shown in Fig. 6a. The thermogram of the scraped-off powders is similar to that of the synthetic manganese hureaulite (Fig. 6b). The dehydration of hureaulite with higher iron ([I) contents is apt to proceed via three steps. [n the case o f hureaufite, (Mno.s3, Feo.47)s H2(P04)4" 3441
4H20, endothermic peaks appear at 286,328, 431, and 980 ~ C, and the weight loss of the corresponding peaks is 6.6, 3.9, 1.5, and 0%, respectively. The exothermic peak is observed around 538 ~ As hureaulite loses 9.8% of its weight due to the dehydration of four water molecules, the dehydration is found to be accomplished under 350 ~ C. The endothermic energy at 980~ causes the fusion of manganese phosphate compounds. On the basis of the foregoing results, samples of synthetic hureaulite were heated at a rate of t0 ~ C min -1 in nitrogen gas and held for 30 min at 280,380, 580,850, or 980 ~ C. The diffraction lines of samples heated at 280 and 380 ~ C are not recognized clearly. X-ray diffraction patterns of samples heated at 580,850 and 980~ prove the presence of anhydrous manganese-iron phosphate compounds, which are presumed to be composed of wolfeite, (Fe, Mn)2(PO4) (OH), hetrosite, (Fe, Mn)PO4, and other unidentified compounds. M6ssbauer spectra and parameters of these pyrolysis products are shown in Fig. 7 and Table III, respectively. The peaks due to the paramag3442
,:........ ......... ...'.. ",'
0 1 2 3 Vetocity ( m m s e c -1)
Figure 7 M6ssbauer spectra of pyrolysis products of hureaulite. Heating temperature (30 min) - (a) 280 ~ C, (b) 380 ~ C, (c) 580 ~ C, (d) 850 ~ C, (e) 980 o C.
netic iron (III) species (isomer shift (I.S.)= 0.48 mm sec -1, quadrupole splitting (Q.S.) = 0.65 mm sec -a) are found in all spectra. It is considered that the iron (III) compound is the oxidation product of iron (II) in hureaulite because the samples were heated in an atmosphere of nitrogen and residual air. The M6ssbauer parameters of phosphate compounds produced by the thermal transformation of hureaulite vary slightly. Characteristic doublet peaks of iron (II) species (I.S. = 1.18mmsec -1, Q.S. = 3.46mmsec -a) are seen in the M6ssbauer spectrum of the sample heated at 580 ~ C through an exothermic process, besides the paramagnetic iron (II) (I.S. -- 1.2 mm sec -1, Q.S. = 2.4 mm sec-1; I.S. = 1.1mmsec -1, Q.S. = 1.9mmsec -1) and iron (III) ( I . S . = 0 . 5 m m s e c -1, Q . S . = 0 . 6 m m sec -1 ) species. Hureaulite is concluded to have a good resistance to thermal deterioration against the dehydration as compared with hopeite  and phosphophyllite , deposited by phosphating in a zinc phosphate bath.
T A B L E I I I M6ssbauer parameters of pyrolysis products of hureaulite, (Mno.53, Feo.47)sH 2(PO4) 4 9 4H20 Heating temp. (~
The variations o f the chemical c o m p o s i t i o n o f p h o s p h a t e d coatings and the physical state o f the interface b e t w e e n substrate and coating in the course of f o r m a t i o n o f manganese p h o s p h a t e coating are revealed by means o f CEMS and TMS. Three different iron (II) species and one iron (III) species are f o u n d in the manganese phosphate coating. Three iron (II) species were identified to be the iron (II)-ion substituted hureaulites o f the three manganese sites o f the octahedral crystal lattice o f manganese hureaulite to the order o f one, two and three sites. The degree o f iron substitution for manganese sites in hureaulite deposited at the steel surface by manganese phosphating depends on the phosphating time, as well as the c o n c e n t r a t i o n o f iron (II) and manganese (II) ions in the phosphating bath. The d e h y d r a t i o n and thermal t r a n s f o r m a t i o n processes o f manganese p h o s p h a t e coatings and synthetic hureaulite were clarified. The D T A curve of the scraped-off sample from manganese p h o s p h a t e coating is similar to that o f manganesee n r i c h e d hureaulite. A characteristic iron (II) species is f o u n d to be p r o d u c e d in the pyrolysis p r o d u c t s of hureaulite h e a t e d to temperatures higher than the e x o t h e r m i c point.
The authors express their thanks to Mr R. Kojima o f the Technical Research L a b o r a t o r y , N i h o n Parkerizing Co Ltd for donating the manganese p h o s p h a t e coatings and his collaboration during the collection of the thermal analysis data.
References 1. G. LORIN, "Phosphating of Metals" (Finishing Publications Ltd, Middlesex, UK, 1974). 2. A. NEUHAUS and M. GEBHARDT, Werkstoffe Korrosion 17 (1966) 567. 3. Y. UJIHIRA and K. NOMURA, Jr. Phys. 41 (1980) C1-392, 4. K. NOMURA, Y. UJIHIRA, Y. MATSUSHITA, R. KOJIMA and Y. SUGAWARA, Nippon Kagaku Kaishi (1980) 1371. 5. K. NOMURA and Y. UJIHIRA, Proceedings of the International Conference on the Applications of the M6ssbauer Effect, Jaipur, India, December, 1981, to be published. 6. F.G. BERRY, J. Chem. Soe. Dalton II (1979) 1736. 7. K. NOMURA, Y. UJIHIRA and R. KOJIMA, Nippon Kagaku Kaishi, (1982) 791. 8. Mem, ibid. (1982) 87. 9. K. NOMURA and Y. UJIHIRA, ibid. (1982) 1352. 10. C. FELDMAN,Phys. Rev. 117 (1960) 455. 11. B. L. CHRISMAN and T.A. TUMOLILLO, Com. puterPhys. Commun. 2 (1971) 322. 12. P.B. MOORE and T. ARAKI, Amer. Mineral. 58 (1973) 302. 13. S. MENCHETTI and C. SABELLI, Acta Cryst. B29 (1973) 2541.
14. M.T. HIRVONEN, Nuel. lnstrum. Meth. 165 (1979) 67. 15. R.J. HILL and J. B. JONES, Amer. Mineral. 61 (1976) 987.