Matrix Dissolution Techniques Applied to Extract and ... · Matrix Dissolution Techniques Applied...

Matrix Dissolution Techniques Applied to Extract and QuantifyPrecipitates from a Microalloyed Steel

JUNFANG LU, J. BARRY WISKEL, OLADIPO OMOTOSO, HANI HENEIN,and DOUGLAS G. IVEY

Microalloyed steels possess good strength and toughness, as well as excellent weldability; theseattributes are necessary for oil and gas pipelines in northern climates. These properties areattributed in part to the presence of nanosized carbide and carbonitride precipitates. Tounderstand the strengthening mechanisms and to optimize the strengthening effects, it isnecessary to quantify the size distribution, volume fraction, and chemical speciation of theseprecipitates. However, characterization techniques suitable for quantifying fine precipitates arelimited because of their fine sizes, wide particle size distributions, and low volume fractions. Inthis article, two matrix dissolution techniques have been developed to extract precipitates from aGrade100 (yield strength of 690 MPa) microalloyed steel. Relatively large volumes of materialcan be analyzed, and statistically significant quantities of precipitates of different sizes arecollected. Transmission electron microscopy (TEM) and X-ray diffraction (XRD) are combinedto analyze the chemical speciation of these precipitates. Rietveld refinement of XRD patterns isused to quantify fully the relative amounts of the precipitates. The size distribution of thenanosized precipitates is quantified using dark-field imaging in the TEM.

DOI: 10.1007/s11661-010-0579-6� The Minerals, Metals & Materials Society and ASM International 2010

I. INTRODUCTION

IT is known that the combination of microalloyingadditions, such as Nb, Ti, and V, and thermomechani-cal-controlled processing (TMCP) can enhance thestrength and ductility of so-called microalloyed steels.The improvement of mechanical properties resultsmainly from the refinement of ferrite grain size, precip-itation, solute strengthening, and dislocation strength-ening,[1–3] of which the most important are grainrefinement and precipitation strengthening. The firstthree can be expressed mathematically by Eq. 1.[3]

ryðMPaÞ ¼ ri þ kyd�1=2 þ

10:8v1=2f

X

!ln

X

6:125� 10�4

� �

þX

kiCi ½1�

where ry is the yield stress, ri is the friction stress of iron,ky is the strengthening coefficient for grain size, d is thegrain diameter in microns, X is the size of precipitates inmicrons, vf is the volume fraction of a given precipitatesize (X), ki is the strengthening coefficient for solute

strengthening of solute i, and Ci is the concentration ofsolute i.Equation 1 does not include a term for dislocation

strengthening because cold deformation is not usedcommonly in hot-rolled or normalized microalloyedsteels.[3]

Different sized precipitates have been identified inmicroalloyed steels, and their precipitation is a sequen-tial process because of solubility differences. Figure 1shows a comparison of the solubility products (e.g., forNbC, Ks = [Nb]equilibium[C]equilibrium = 10(A–B/T), whereKs is the solubility product, T is temperature in Kelvin,and A and B are constants for a given system) formicroalloyed carbides and nitrides. The solubility ofprecipitates establishes guidelines to identify differentphases. It can be observed that TiN is the least soluble ofthe precipitates in austenite. TiN should precipitate outat higher temperatures during earlier processing stagesand can be several microns in size. NbC has a muchhigher solubility and will precipitate out at lowertemperatures during the subsequent processing stageswith a much smaller size. The solubilities of VC and VN,especially VC, are much higher than those of the otheralloying element carbides/nitrides in austenite andferrite. Molybdenum carbide has a similar solubility toVC.[4] Because of their high solubility, Mo and Vcarbides are expected to precipitate last and have thefinest particle size. Therefore, precipitates can be clas-sified into different groups according to precipitate sizeand their precipitation sequence.Precipitates of different sizes contribute to each of

these strengthening mechanisms in various ways. Tomaintain a fine austenite grain size prior to transforma-tion, particles that precipitate in austenite or particles

JUNFANG LU, formerly PhD Graduate Student with the Depart-ment of Chemical and Materials Engineering, University of AlbertaEdmonton, Alberta T6G 2V4, Canada, is now Engineer with EnbridgePipelines Inc., Edmonton, Alberta T5J 3N7, Canada. J. BARRYWISKEL, Research Service Officer, and HANI HENEIN andDOUGLAS G. IVEY, Professors, are with the Department of Chemicaland Materials Engineering, University of Alberta. Contact e-mail:[email protected] OLADIPO OMOTOSO, Engineer, is withSuncor Energy Inc., Fort McMurray, Alberta T9H 3E3, Canada.

Manuscript submitted March 16, 2010.Article published online December 23, 2010

METALLURGICAL AND MATERIALS TRANSACTIONS A VOLUME 42A, JULY 2011—1767

that will precipitate during hot rolling are required.These precipitates, typically carbides, nitrides and/orcarbonitrides of Nb and Ti, are >20 nm in size.[5,6] Toproduce precipitation strengthening, fine particles(<20 nm), especially those smaller than 5 nm, areneeded.[7] The finest nanoprecipitates are particularlyimportant for precipitation strengthening.

Precipitation strengthening increases as the volumefraction of the precipitates increases and the precipitatesize decreases.[1] For a given volume fraction, refinementof particle size means a higher number density ofprecipitates, which means a higher number of interac-tions between precipitates and the dislocations in theferrite matrix. Thus, a significant increase in yield stresscan be produced. Therefore, refining precipitate size andincreasing the volume fraction of the nanoprecipitatesare important issues in the steel industry. However,quantitative determination of the volume fraction of thenanoprecipitates is challenging because of the smallparticle sizes, wide particle size distribution, and lowvolume fraction of precipitates. Another challenge isthat Ti and Nb carbonitrides have the same crystalstructure (NaCl-type), with similar lattice parameters. Itis difficult to differentiate the precipitates solely on achemical composition basis, making the problem morecomplicated.

Microscopic methods are used commonly for charac-terizing precipitates. Both optical microscopy and scan-ning electron microscopy (SEM) have limited resolution.Transmission electron microscopy (TEM) has superiorimaging resolution and spatial resolution for energydispersive X-ray (EDX) analysis. However, TEM anal-ysis does have limitations because of the small specimensused. More recently, atomic force microscopy (AFM)[8]

and atomprobe field-ionmicroscopy (APFIM) have beenused.[9,10] Similar to TEM, both AFM and APFIM canprovide high resolution images. The advantages of AFMare its lower cost and ease of operation compared withTEM. APFIM is not as common but has advantages for

nanoprecipitate investigation. High-resolution imagesare obtained by ionization of atoms from the specimen.[9]

The chemistry of the specimen can be determinedaccurately on a single particle as small as 1 nm indiameter.[9,10] The disadvantage of APFIM is that thevolume of material analyzed is much smaller than that inTEM. Sample preparation for APFIM is also complexand time consuming with only a small volume analyzedfor each measurement. The sampling volume for AFM isalso small.With the previously discussed microscopic techniques,

microstructural features of the steel matrix and precip-itates have been examined. Qualitative or semiquantita-tive methods may have been sufficient in helping toincrease steel strength to date; however, future improve-ments will need to be more quantitative. To improvesteel mechanical properties effectively for the futuredevelopment of new steels with higher strength, bettertoughness, and better weldability, such as X100, X120(X refers to pipeline grade and 100/120 refers to thespecified minimum yield strength in ksi), and beyond, amore quantitative understanding of the strengtheningmechanisms is required. Based on the quantitativeinformation, the relative importance and contributionof different strengthening mechanisms can be deter-mined. However, this quantitative information is lack-ing and microscopic methods are not completelysatisfactory for quantifying precipitate size, size distri-bution, and volume fraction of precipitates in steelsbecause of the low volume of precipitates.Based on previous work by the authors, matrix

dissolution has been shown to be a successful techniqueto quantify precipitates from samples that are morerepresentative of the steel strip. Relatively large volumesof material are analyzed, so that statistically significantquantities of precipitates of different sizes are collected.Matrix dissolution can be done either chemically orelectrolytically to extract the carbonitrides.[11]

For electrolytic dissolution, it is possible to extractprecipitates selectively by composition because of thedifferent dissolution behavior of the various precipitatesand the steel matrix. The ferrite matrix is dissolved atlow anodic potentials, while precipitates are more nobleand are dissolved at higher potentials. Some precipitatesare more noble than others, thereby allowing selectiveextraction.It is noted that matrix dissolution methods have been

successful in extracting Ti/Nb/Mo/V carbonitrides, bothchemically and electrolytically, because of their insolu-bility in HCl acid and selected electrolytes. This tech-nique may not be applicable to extract mixed Fe-Cr-Mnnanocarbides as reported by Chakraborty et al.[12,13]

They dissolve partially or are retained or embeddedduring refinement of the bainitic-ferrite sheaf. Thesemixed carbides may be soluble in HCl acid or anelectrolyte.In this article, both electrolytic and chemical dissolu-

tion methods are used and the results are compared.After the steel is dissolved, precipitates are separatedfrom the solution by centrifuging. The extracted precip-itates are specifically identified using TEM and X-raydiffraction (XRD) without interference from the matrix.

Fig. 1—Equilibrium solubility products for microalloyed carbidesand nitrides in austenite and ferrite.[3]

1768—VOLUME 42A, JULY 2011 METALLURGICAL AND MATERIALS TRANSACTIONS A

Precipitate size is determined by TEM imaging and thechemistry is determined through TEM X-ray micro-analysis. XRD is also used to provide valuable crystal-lographic information. Rietveld refinement of XRDpatterns is used for refining precipitate structures and toobtain their relative abundance, making it possible todetermine the volume fraction of precipitates in thesteel.

II. EXPERIMENTAL PROCEDURE

Grade100 microalloyed steel, with a thickness of8 mm and supplied by Evraz Inc. NA (Portland, OR),was used in this study. This is a high-strength, low-carbon, low-alloy structural steel with a specified min-imum yield strength of 100 ksi (690 MPa). TheGrade100 designation is the strength equivalent ofCanadian Standards Association (CSA) Grade 690,which is the designation suggested by the CSA. Thissteel is produced through thermomechanical controlledprocessing (TMCP). The chemical composition of rele-vant elements is given in Table I. Manganese is one ofthe principal alloying agents in most steels. Both Mnand Mo provide solid-solution strengthening and sup-press the austenite to ferrite phase transformation,resulting in the formation of lower transformationtemperature products (such as acicular/bainitic ferrite)and finer precipitate dispersion.[14,15] Manganese is alsoadded in steels to combine with sulfur to form manga-nese sulfide (MnS).[16] Small amounts of elements suchas Ti, Nb, and V are added to this steel to increase thestrength. Aluminum is not usually considered as amicroalloying element. It is added primarily as adeoxidizer to the steel[17] and works in conjunction withsilicon.

Both electrolytic and chemical dissolution were usedto dissolve the steel matrix and extract precipitates.Samples approximately 1.5 to 2 cm3 in volume, takenthrough the plate thickness, were used in both dissolu-tion methods. An analytical balance with an accuracy of0.1 mg was used to measure the weight of the samplebefore dissolution and the weight of the extractedresidues.

Various acids, such as HCl, H2SO4, and HNO3, havebeen reported in the literature for extracting differenttypes of particles. Only those phases stable in acids, suchas silica, alumina,[18] and several carbides and nitrides,have been extracted successfully. ASTM E194-90[19] is astandard test method to determine the acid-insolublecontent of iron powders. In this standard, a dilute HClsolution is used. Insoluble matter, such as carbides,silica, insoluble silicates, alumina, clays, and otherrefractory materials, can be collected. Data are unavail-able for the solubility of TiC, TiN, VC, and VN in acids.However, it is reported that NbC and NbN are insoluble

in any acid.[20] Because of their similar crystal structures(all have NaCl-type structures), it is reasonable to expectthat these microalloyed carbides and nitrides will beinsoluble in HCl.To define the appropriate conditions for electrolytic

dissolution of the steel matrix, potentiodynamic curvesfor the Grade100 steel and several binary carbides andnitrides were obtained using a Gamry Potentiostat/Galvanostat system (Gamry PC4/750, Gamry Instru-ments, Warminster, PA). The electrolyte employed was10 pct acetylacetone (AA) solution,[21] which consists of10 pct acetylacetone, 1 pct tetramethylammonium chlo-ride and methanol. During each potentiodynamic scan,the steel sample, pure carbide, or pure nitride acted asthe anode, with platinum sheet as the cathode. Using thesame Gamry system, a static potential (300 mV vssaturated calomel electrode [SCE]) was applied duringelectrolytic dissolution to dissolve the matrix and toleave behind the precipitates. For chemical dissolution,an HCl acid solution (1:1 mixture by volume of HClacid, with a specific gravity 1.19, and distilled water) wasused at 338 K to 343 K (65 �C to 70 �C). Henceforth,dissolution using 10 pct AA will be referred to aselectrolytic dissolution, and dissolution using the HClsolution will be referred to as chemical dissolution.After the steel matrix was dissolved, a Sorvall RC-6

super speed centrifuge (Mandel, Guelph, ON, Canada)made by Mandel, was used to separate solid particlesfrom the solution by rotating the material rapidly at upto 40,000 RCF (relative centrifugal force) at 277 K(4 �C). The centrifuging process was repeated severaltimes to clean the precipitates. Inductively coupledplasma spectroscopy (ICP), with a Perkin Elmer Elan6000 ICP-MS (Perkin Elmer, Waltham, MA), was usedto analyze the concentration of the elements in thesupernatant extracted by chemical dissolution, as ameans of performing a mass balance. Calibration wasdone using four calibration levels (four-point calibrationcurve) for each element. The solution to be analyzed wasdiluted to keep the analyte concentration within thelinear working range.For phase identification, diffraction data were col-

lected with a Bruker D8 Advance X-ray diffractometer(Bruker AXS Inc., Madison, WI), equipped with a CoX-ray source and a VANTEC-1 linear detector. Twotechniques were used for TEM sample preparation: oneusing precipitates extracted during matrix dissolutionand the other using precipitates produced from carbonextraction replicas. For the first case, a fraction of theprecipitates extracted using electrolytic or chemicaldissolution was dispersed in ethanol. A drop of thesuspension was deposited on a carbon-coated, 300-meshCu grid, and the solvent was allowed to evaporate,leaving behind the particles. These samples were used toidentify specific types of precipitates. The compositionsof the precipitates were determined by EDX analysis in

Table I. Chemical Composition of Grade100 Microalloyed Steel (Weight Percent)

Element C Mn Si N Cu + Ni + Cr Mo Nb Ti V Al Ca

Wt pct 0.080 1.800 0.244 0.011 0.877 0.301 0.094 0.06 0.047 0.05 0.005


the TEM, where a standard-less approach was used. Forthe second case, a thin carbon film was evaporated ontothe surface of a polished/etched steel metallographicsample. TEM samples were taken from regions throughthe plate thickness, although the midthickness andsurface regions were avoided. The carbon film wasstripped from the surface and extracted precipitatesfrom the surface region of the ferrite matrix. Dark-field(DF) imaging was used to characterize the size distri-bution of nanoprecipitates extracted from the ferritematrix onto carbon replicas supported on Cu grids. Thecompositions of the precipitates in the carbon extractionreplicas were determined by EDX analysis in the TEMas well. The first technique (matrix dissolution) pro-duced many more particles than the second technique(carbon extraction replicas), which will be discussedsubsequently. The samples were examined in a JEOL2010 TEM (JEOL Ltd., Tokyo, Japan) equipped withan EDX detector operating at 200 kV.

TOPAS Academic software (Bruker AXS Inc.,Madison, WI) was used for Rietveld analysis of theprecipitate structures and to determine their composi-tion. The technique uses a nonlinear least squaresmethod to refine instrument and structural parameterscontributing to a calculated diffraction pattern until agood match is generated.[22,23] Given that precipitateshave similar structures and lattice parameters, theelement occupancies were fixed based on the precipitatechemical composition obtained by EDX analysis. Thescale factor and lattice parameters (with tight restraints)were refined to reduce strong parameter correlations.The phase fractions were derived from Eq. [2].[24]

wa ¼SFa MZVð ÞPj

SFj MZVð Þj½2�

where wa is the relative weight fraction of phase a in amixture of j phases, SF is a refined scale factor and isproportional to the number of unit cells of phase a in thespecimen, M is the mass of the molecular formula, Z isthe number of formula units per unit cell, and V is thevolume of the unit cell.

III. RESULTS

A. Polarization Behavior of Grade100 Steel and BinaryPrecipitate Materials

Polarization curves using electrolytic dissolution forthe Grade100 steel (ferrite matrix) and several binarycarbides and nitrides are shown in Figure 2. Clearly, thesteel matrix is the most soluble followed by TiN, TiC,and then NbC and NbN. The matrix can be dissolvedselectively, at potentials<1000 mV vs SCE. In addition,TiN (the largest of the precipitates in the steel) iselectrolytically dissolved preferentially relative to theother carbides/nitrides at potentials<1500 mV vs SCE,leaving behind the other compounds. If a high potential,i.e., 1500 mV, is supplied within the current range of theequipment, then smaller precipitates can be extractedfrom the ferrite matrix and TiN selectively dissolved.

However, this has been shown to be impractical for theGamry system used in this work. If a significant amountof precipitates is to be extracted, then the total surfacearea of the steel to be dissolved needs to be several cm2,which means the current will exceed the current range ofthe system. This issue could be overcome by the use ofseveral power supplies and smaller specimens.

B. Mass Balance of the Supernatant and the ExtractedResidue

The experimental extraction yield, i.e., total wt pct ofthe residue relative to the amount of steel beingdissolved, by two dissolution methods is shown inTable II.For HCl dissolution, ICP analysis was performed on

the centrifuged solution after the precipitates had beenextracted from the HCl solution. By comparing thechemical composition of the steel with the ICP analysis ofthe solution, the amount of each element extracted fromthe steel as part of the residue was obtained, as shown inTable III. Most of the Si in the steel was present in theresidue as SiO2, which was verified by the following TEManalysis. Most of the Nb in the steel was in precipitateform. There are lower yields for Ti, Mo, and V, especiallyfor Mo and V, which means that most of the Mo and Vwere dissolved in solid solution in the ferrite matrix.

C. Preliminary XRD Analysis of the Extracted Residues

Figure 3 shows XRD patterns for residues extractedfrom the Grade100 steel; samples were obtained throughelectrolytic and chemical dissolution, respectively. Themain difference for the two matrix dissolution tech-niques is the presence of a broad peak for chemical

Fig. 2—Polarization curves for Grade100 steel and binary precipi-tates in the 10 pct AA solution.

Table II. Experimental Extraction Yield of Residue

from Two Dissolution Methods

Dissolution methodElectrolytic(10 pct AA)

Chemical(HCl)

Experimental extractionyield (wt pct)

0.361 pct 0.792 pct


dissolution between the 2h angles of 20 deg and 30 deg,which is because of the presence of an amorphous phase,identified by TEM analysis as silica. The silica phase isdiscussed in more detail in subsequent sections. Thebroad peak was not found in the XRD pattern forresidues extracted by electrolytic dissolution, which isadvantageous for profile fitting in Rietveld analysis. Theabsence of the broad amorphous peak in the sampleallows for a better signal-to-noise ratio and peakresolution in the diffraction pattern, making Rietveldrefinement less error prone. However, on the negativeside, electrolytic dissolution is much more time consum-ing than chemical dissolution.

A preliminary analysis of the diffraction patternsshows the apparent presence of two crystalline phaseswith similar structures; it will be shown subsequentlythat more phases are actually present. One set of peaks(indicated by circles in Figure 3) seems to representNbC-based precipitates (larger d spacings), and theother set (squares in Figure 3) corresponds to TiN-based precipitates. Both types of precipitates have aNaCl-type (face-centered cubic [fcc]) structure with anFm-3m space group. There are intensity differences,particularly for the {111} and {200} peaks among thetwo patterns, which is because of sample orientationeffects during XRD measurements.

Because the crystal structure (NaCl-type) and latticeparameters of the precipitates are similar and the latticeparameters vary with composition, it is difficult todifferentiate between the various carbides, nitrides, andcarbonitrides.

D. TEM Analysis of Nanoprecipitates Extractedby Carbon Replicas

Precipitates, especially the nanoprecipitates, can beextracted via carbon replicas. Because of the difficulty in

analyzing fine precipitates, the size distribution of thenanoprecipitates was investigated using dark-field imag-ing. Figure 4(a) shows an example of a TEM bright-field(BF) image from a Grade100 carbon replica. The insetshows the corresponding selected area diffraction (SAD)pattern from the field of view. Several intermittent ringsare visible and correspond to fine precipitates with aNaCl-type crystal structure. Figure 4(b) shows a DFimage, taken using part of the {111} and {200}diffraction rings. Several DF images were taken byvarying the position of the objective aperture around thediffraction rings, so that all particles in the field of viewwere selected. Image processing software (ImageTool)was used to analyze the DF images and to quantifythe size distribution of the nanosized precipitates.Figure 4(c) shows an EDX spectrum from some of thenanosized precipitates. They are Nb/Mo-rich withsmaller amounts Ti and V. Although C is present inthe precipitates, the large C peak arises primarily fromthe carbon film in the replica. The Cu peaks are anartifact arising from the Cu support grid and the smallFe peak is residual matrix material. The relativefrequency vs diameter of nanoprecipitates in theGrade100 steel is shown in Figure 4(d). More than2000 nanoprecipitates (mainly £10 nm) were countedfor the size distribution and chemistry analysis. Therelative frequency (n/N) is defined as the ratio of thenumber of particles (n) within a given size range tothe total number of particles counted in that region (N).The largest number of precipitates is in the 3 to 5 nmsize range. From Figure 4(d), the diameter with thelargest number distribution is approximately 4.5 nm.

E. Characterization of Precipitates Extracted by MatrixDissolution

The size and chemistry of the precipitates extracted bymatrix dissolution were characterized by TEM imagingand EDX microanalysis. Compared with carbon extrac-tion replica methods, approximately 10,000 times moreprecipitates are extracted by matrix dissolution, makingphase quantification possible. This estimation is basedon the thickness of steel-contributing precipitates to thecarbon replicas and the amount of steel dissolved duringmatrix dissolution.

1. Large and intermediate-size precipitates extractedby chemical and electrolytic dissolutionFigure 5 shows a TEM BF image of a large cuboidal

precipitate approximately 3 lm in size. It was extractedby chemical dissolution. EDX analysis indicates that it isTi-rich with a small amount of Nb. It is likely that thelarge cuboidal precipitates are formed at high temper-ature, in the liquid, or during or soon after casting.

Table III. ICP Analysis of the Supernatant After Chemical Dissolution

Grade100 Si Nb Ti Mo V

Steel chemistry (wt pct) 0.244 0.094 0.06 0.301 0.047Supernatant chemistry (wt pct) 0.021 0.003 0.014 0.257 0.045Extraction yield of precipitates 91.6 pct 97.2 pct 76.3 pct 14.5 pct 5.0 pct

Fig. 3—XRD patterns for residues extracted from Grade100 steel byelectrolytic (10 pct AA) and chemical dissolution (HCl).


They are too large to affect austenite grain size refine-ment during reheating. As such, it is desirable to avoidthe formation of this type of precipitate.

Some epitaxially nucleated precipitates were extractedby both chemical and electrolytic dissolution. Figure 6shows an example of a TEM BF image, EDX spectra,and SAD patterns from the Nb-rich precipitatesattached to a Ti-rich precipitate, extracted by electro-lytic dissolution. Because TiN has a lower solubility inFe (either austenite or ferrite) than NbC, NbN, or Nbcarbonitrides, it precipitates first in the steel on coolingand can act as nucleation sites for Nb-rich precipitatesas shown in Figure 6(a). Figures 6(b) and (c) show EDXspectra of the Ti-rich precipitate and one of the Nb-richprecipitates in Figure 6(a). The source of the C, Cu, andFe is the same as explained previously. An SAD pattern,with a zone axis close to [114], of the overlapping Ti-richprecipitate and Nb-rich precipitate is shown inFigure 6(d). Figure 6(e) shows an SAD pattern of the

Ti-rich precipitate in Figure 6(a). The (262) reflection inFigure 6(d), which is indicated by an arrow, shows spotsplitting caused by the slightly different lattice param-eters for the two phases. The Ti-rich precipitates have asmaller lattice parameter than the Nb-rich precipitates.Therefore, spots A (smaller d spacing) and B (largerd spacing) in Figure 6(d) come from the Ti-rich and Nb-rich precipitates, respectively. The coarse, Ti-richnitrides (cuboidal in shape) are formed during or soonafter casting (as discussed previously), whereas the Nb-rich precipitates form in the austenite phase with manygrowing epitaxially on Ti-rich precipitates. This phe-nomenon has been reported previously. Epitaxialgrowth is favored as the decrease in interfacial energybetween the Nb-rich precipitate and preexisting TiNprecipitate is much larger than the increase in strainenergy from the small lattice misfit (~2 pct).[4,25]

Another type of intermediate-size precipitate(~100 nm in size) was observed in the Grade100 steel.

0

100

200

300

0 2 4 6 8 10 12 14 16 18 20

Inte

nsit

y

Energy (keV)

Grade 100C

O

Cu Nb

Ti

Fe

Cu

Cu Nb

Nb Mo

Mo

Ti V

0.00

0.05

0.10

0.15

0.20

-1 1 - 2

2 - 3

3 - 4

4 - 5

5 - 6

6 - 7

7 - 8

8 - 9

9 - 10

10 - 12

12 - 14

Precipitate size (nm)

Rel

ativ

e fr

eque

ncy

of n

ano

prec

ipit

ates

(n/

N)

Grade 100

200

{111} {220}

(a) (b)

(c) (d)

Fig. 4—(a) TEM BF image of nanoprecipitates extracted via carbon replicas. (b) TEM DF image of same region as shown in (a). (c) EDX spec-trum from several nanoprecipitates. (d) Size distribution of nanoprecipitates.


A representative BF image and an EDX spectrum areshown in Figure 7. EDX analysis indicates that it hassimilar amounts of Ti and Nb (~50 at. pct Ti and~50 at. pct Nb).

2. TEM analysis of nanoprecipitates extractedby matrix dissolution

Figure 8 shows TEM images of extracted nanopre-cipitates from the Grade100 steel by chemical dissolu-tion. The finest precipitates extracted are less than 5 nmin size (Figure 8(b)). EDX analysis (Figure 8(c)) revealsthat the particles are Nb- and Mo-rich, and they containsmall amounts of Ti and V, which confirms the resultsfrom the extraction replicas in the previous section. Anamorphous Si-O phase (Figures 8(d) and (e)), viewed asspherical particles in Figure 8(a), was also present in theresidue. The Si and O peaks in Figure 8(c) are from thisSi-O phase. This finding is consistent with the ICPanalysis of the supernatant where most of the Si in thesteel matrix was present in the residue. The amorphousSi-O phase is expected to be SiO2, which was confirmedby comparing the d spacing of the broad ring inFigure 8(e) with the most intense peak for cristobalite.To determine whether the source of O for the Si-Oparticles in the residue was the HCl solution (i.e.,dissolved oxygen), N2 was bubbled through the solutionboth prior to and during dissolution. SiO2 was stillextracted in the residue in the same quantities after N2

treatment, so the source of O was likely oxide/silicateinclusions in the steel or dissolved Si in the ferritematrix. During steel dissolution, the inclusions aredissolved in the HCl solution and the dissolved Sicombines with dissolved O to precipitate out as theamorphous Si-O phase.

Nanoprecipitates (£10 nm) can also be extracted byelectrolytic dissolution. Figure 9 shows a TEM BF

image and EDX spectrum of electrolytically extractednanoprecipitates from the Grade100 steel. The nano-precipitates, as expected, are similar in size and inchemistry compared with those extracted by chemicaldissolution. EDX analysis shows that they are Nb/Mo-rich with some Ti and V. The Si peak is much lessintense than that for the chemically extracted residues,because of reduced formation of the amorphous Si-Ophase. This is a clear advantage of electrolytic disso-lution. However, on the negative side, electrolyticdissolution is more time consuming. The precipitateyield was only approximately 10 pct of that generatedby chemical dissolution for the same amount ofextraction time. As such, the sample size for XRDanalysis was significantly smaller for electrolyticallyextracted residues. In addition, it takes approximatelythree times longer to clean the electrolytically extractedresidues.From the preceding TEM imaging and EDX micro-

analysis, different size precipitates with various compo-sitions can be extracted successfully by matrixdissolution using both chemical and electrolytic meth-ods. The precipitates extracted by the different methodsare similar in size and composition and can be catego-rized into five groups accordingly. Precipitate groupingis similar to that observed by other researchers for thesame grade of steel.[6] More than 70 precipitates greaterthan 10 nm in diameter and more than 2000 nanopre-cipitates less than or equal to 10 nm in diameter werecharacterized. Table IV summarizes the precipitateinformation from the Grade100 steel. The transitionelement compositions represent average values mea-sured for precipitates of a particular size. The C and Ncompositions could not be quantified by EDX analysis,because of the presence of the carbon film and overlapof the N K peak with the Ti L peaks. The large Ti-rich

0

500

1000

1500

2000

0 2 4 6 8 10 12 14 16 18 20

Inte

nsit

y

Energy (keV)

Nb

Ti

Ti

Fe

Cu

Cu Nb NbN

(a) (b)

Fig. 5—(a) TEM BF image of a large Ti-rich precipitate extracted by chemical dissolution using HCl. (b) EDX spectrum from the Ti-rich precip-itate in (a).


0

200

400

600

800

1000

1200

Inte

nsit

y

Energy (keV)

Nb

Ti

Cu

Cu NbNb

Ti

CuO

0

200

400

600

0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20

Inte

nsit

y

Energy (keV)

Nb

Ti

Cu

Cu

Nb

Nb

Ti

Cu

C

Fe

1: Ti-rich

2: Nb-rich

3: Nb-rich

ZA close to ]141[−

ZA close to ]141[−

000

220

1 3 1

311

220

000

1 3 1311

2 6 2

A B

(a)

(b)

(d) (e)

(c)

Fig. 6—(a) TEM BF image of Nb-rich precipitates attached to a Ti-rich precipitate, extracted by electrolytic dissolution using 10 pct AA.(b) EDX spectrum from Ti-rich precipitate in (a). (c) EDX spectrum from one of the Nb-rich precipitates in (a). (d) SAD pattern from overlap-ping Ti-rich and Nb-rich precipitates (interface 1/2) in (a). (e) SAD pattern from Ti-rich precipitate in (a).


precipitates (300 to 3000 nm) are formed at highertemperatures (during or soon after casting) and areexpected to contain little or no C, so the composition islisted as Ti(Nb)N. The intermediate-size Ti-rich (100 to300 nm), Nb-rich and Ti/Nb-rich (100 to 200 nm)precipitates are formed during the rolling process. Thesecontain both C and N. To simplify the calculations (atleast initially), the C and N amounts are assumed to besimilar. (The actual composition could vary somewhatfrom this simplifying assumption.) Nanoprecipitates(£10 nm, Nb/Mo-rich) are formed at much lowertemperatures (during or after coiling). It is expectedthat they are primarily carbides. The expectation andassumptions are based on precipitate solubilities.[3]

Matrix dissolution and carbon replicas yield consistentresults for precipitate size and chemistry; in additioncarbon replicas help to validate the matrix dissolutiontechnique.

3. Rietveld refinement of the XRD dataA detailed analysis of the XRD patterns was

obtained via Rietveld refinement. Figure 10 shows theprofile fitting of the XRD pattern for residues obtainedfrom the Grade100 steel by electrolytic dissolution.Five phases are superimposed on the observed diffrac-tion pattern. Figure 10(a) shows the overall profilefitting and the difference between the calculated andobserved spectrum. Figures 10(b) through (f) show thecalculated diffraction patterns for Ti0.9Nb0.1N, Ti0.77-Nb0.23C0.5N0.5, Ti0.5Nb0.5C0.5N0.5, Nb0.7Ti0.3C0.5N0.5,and Nb0.48Mo0.28Ti0.21V0.03C, respectively, as indi-cated by arrows. These precipitate phases were deter-mined through TEM analysis in Table IV. The tickmarks at the bottom of the figures indicate the peakpositions for the previously mentioned phases. Thelattice parameters for the five types of precipitates aresimilar and are listed in Table V. The smallest

precipitates (3 to 5 nm in size) are attributed toNb0.48Mo0.28Ti0.21V0.03C.Rietveld refinement was also carried out on XRD

patterns from residues produced by chemical dissolu-tion. The profile fitting of the XRD pattern for theresidues from chemical dissolution, as shown inFigure 11, is slightly different from that by electrolyticdissolution. Because nanoprecipitates (mostly £10 nm)with chemistry Nb0.48Mo0.28Ti0.21V0.03C have broadpeaks, there is a strong correlation (peak overlap) withamorphous SiO2, i.e., the nanoprecipitate information iscontained within the amorphous SiO2 information.Therefore, the nanoprecipitate phase and amorphousSiO2 are considered together as one phase in theRietveld refinement. As such, the amount of nanopre-cipitates is not determined directly from the Rietveldrefinement. SiO2 was used as an internal standard, andthe resulting residual amorphous content is ascribed tothe nanoprecipitates. The amount of SiO2 is determinedfrom the Si content in the residue, which was obtainedindirectly from the total amount of Si in the steel minusthe amount in the supernatant (from ICP analysis;Table III). To apply Rietveld refinement to the amor-phous SiO2 structure, a crystalline form of silica,cristobalite, is used to approximate the amorphousphase. Cristobalite has a tetragonal structure(a = 0.49732 nm, c = 0.69236 nm),[26] with the samecomposition and similar d spacings to the major broaddiffraction rings for amorphous SiO2. Figure 11(a)shows the overall profile fitting and the differencebetween the calculated and observed spectrum.Figures 11(b) through (f) show the calculated diffractionpatterns (indicated by arrows) for Ti0.9Nb0.1N, Ti0.77-Nb0.23C0.5N0.5, Ti0.5Nb0.5C0.5N0.5, Nb0.7Ti0.3C0.5N0.5,and the nanoprecipitates + amorphous SiO2. The tickmarks in the bottom of the figures indicate the d-spacingpositions for the previously mentioned phases.

0

200

400

600

800

0 2 4 6 8 10 12 14 16 18 20

Inte

nsit

y

Energy (KeV)

C

Nb

Ti

TiFe

Cu

Cu

Nb

Nb

(a) (b)

Fig. 7—(a) TEM BF image of an intermediate-sized Ti/Nb-rich precipitate extracted by chemical dissolution using HCl. (b) EDX spectrum fromthe precipitate shown in (a).


The goodness of fit can be evaluated from thedifference plot between the observed and calculated dataor by the Bragg R-index (RBragg). The difference plot,yi(obs)-yi(calc), is the gray line at the bottom of the fittedprofiles (Figures 10 and 11). RBragg implies a comparison

of integrated intensities similar to single crystal refine-ment. RBragg and the lattice parameters for the differentphases in the residues from electrolytic and chemicaldissolution are shown in Table V. The lattice parameterswere calculated from the d spacings determined from the

0

100

200

300

400

Inte

nsit

y

Energy (keV)

C

O

Cu

Si Nb

Ti Ti Fe

Cu

Cu Nb

NbCa Mo

Mo

Mo Fe V0

400

800

1200

0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10

Inte

nsit

y

Energy (keV)

C

O

Cu

Si

Cu

Cu

Nb/Mo rich

Nb/Mo rich

Si-O

(a) (b)

(c) (d)

(e)

Fig. 8—(a and b) TEM BF image of fine Nb/Mo-rich precipitates extracted by chemical dissolution using HCl. (c) EDX spectrum fromNb/Mo-rich precipitates shown in (b). (d) EDX spectrum from Si-O phase shown in (a). (e) SAD of Si-O phase shown in (a), indicatingamorphous structure.


XRD patterns. A comparison of the lattice parametersfor the precipitates extracted by the two differentdissolution methods shows good agreement, which isexpected as the precipitates are from the same Grade100steel; only the extraction methods are different.Ti0.9Nb0.1N has the smallest lattice parameter. Thelattice parameter increases with increasing Nb contentand decreases with increasing Mo and V content.

Calculations were done to determine whether thelattice parameters for the Ti/Nb carbonitrides inTable V correlated with the compositions obtained fromTEM analysis (Table IV). Lattice parameters for binaryTi and Nb carbides and nitrides are shown in Table VI;these were obtained from the International Center forDiffraction Data database.[26]

Because these binary phases all have the same crystalstructure with similar lattice parameters, it is reasonableto assume that the lattice parameters for the ternary

phases (Ti/Nb carbides, Ti/Nb nitrides, or Ti/Nbcarbonitrides) can be estimated by linear interpolationbetween the appropriate binary phases. For example,the lattice parameter for TixNb1–xN can be calculated bylinearly interpolating between the lattice parameters forTiN and NbN for the desired value of x. This processcan be used in reverse, i.e., the composition of aparticular precipitate can be determined if its latticeparameter is known. The calculations were done usingthe lattice parameters for the first four precipitates inTable V; the fifth precipitate was excluded because ofthe compositional complexity (four transition metalcomponents). Two assumptions were made. The firstprecipitate in Table IV was assumed to be a nitride (nocarbon) to determine the Ti and Nb compositions. Thisassumption is reasonable as it is the largest precipitate,forms at the highest temperature, and, as such, isexpected to contain little or no C. The other threeprecipitates were assumed to be carbonitrides. The Tiand Nb compositions, which were obtained from aTEM analysis, were taken as correct, so that the C andN compositions could be determined. Note that inTable IV, the C and N amounts were assumed to be 0.5in the precipitate formulae because the C and Ncompositions could not be determined from EDXanalysis. The resultant precipitate compositions areshown in Table VII. The first precipitate (Ti/Nb nitride)in the table shows good agreement with the compositionobtained from TEM EDX analysis (Table IV), i.e., thedifference is within one standard deviation. For theother three precipitates, the relative N and C composi-tions show the correct trend, i.e., the C amount increasesas the precipitates become more Nb rich (Ti deficient).As mentioned previously, Al is added primarily as a

deoxidizer in steel and works in conjunction with silicon.Aluminum is normally dissolved in austenite at high

0

100

200

300

400

0 2 4 6 8 10 12 14 16 18 20

Inte

nsit

y

Energy (keV)

Si

Nb

Ti

Fe

Cu

Cu

Nb

Nb Ca

Mo

MoFe V

Ti

Mo Nb/Mo rich

(a) (b)

Fig. 9—(a) TEM BF image of fine Nb/Mo-rich precipitates extracted by electrolytic dissolution using 10 pct AA. (b) EDX spectrum fromprecipitates shown in (a).

Table IV. Classification of Precipitates Accordingto Composition and Size Range Determined from TEM

Analysis

PrecipitateChemistry

Standard Deviationfor Composition

in Atomic Fraction

SizeRange(nm)

Ti0.9Nb0.1N Ti 0.05 300 to 3000Nb 0.05

Ti0.77Nb0.23C0.5N0.5 Ti 0.05 100 to 3000Nb 0.05

Ti0.5Nb0.5C0.5N0.5 Ti 0.07 100 to 200Nb 0.07

Nb0.7Ti0.3C0.5N0.5 Ti 0.05 100 to 200Nb 0.05

Nb0.48Mo0.28Ti0.21V0.03C Ti 0.05 <10Nb 0.05


temperatures prior to rolling. The nitride phase (AlN) hasa low solubility product in steel, and it is thermodynam-ically stable at lower temperatures.[7,16] However, AlNhas a high nucleation barrier because of its hexagonalcrystal structure, i.e., a different morphology from theother FCC nitrides.[16,27] AlN nucleates with somedifficulty in steel unless precipitation is enhanced bythermal or mechanical treatments.[17] In addition, TiNhas an even lower solubility product than AlN. Thermo-dynamic calculations and experimental work have shownthat AlN is not present in this Grade100 steel.[4,11]

In addition to precipitates, some Al- and Ca-contain-ing inclusions were observed by optical microscopy or

SEM. They were not studied in detail in this article, asthey do not contribute to the strength.Manganese was not detected in the precipitates. This is

consistent with the previous statement that Mn primarilyprovides solid solution strengthening.[15] It should benoted that some Mn combines with sulfur to form MnS.However, as with Fe3C, MnS is unstable in acid. BothFe3C and MnS cannot be extracted by acid solutions, asreported by Kanazawa et al.[28] and Bandi.[29]

Iron has been found in small Nb/Mo/V/Ti carbides,which indicates that the precipitates may have nonequi-librium compositions.[4] The addition of iron to theNaCl-type lattice of transition metal carbonitrides

908580757065605550454035

9,0008,0007,0006,0005,0004,0003,0002,0001,000

0-1,000-2,000

Ti0.77Nb0.23C0.5N0.5

908580757065605550454035

9,0008,0007,0006,0005,0004,0003,0002,0001,000

0-1,000-2,000

Obs

Diff

Calc

908580757065605550454035

9,0008,0007,0006,0005,0004,0003,0002,0001,000

0-1,000-2,000

Ti0.9Nb0.1N

(a)

(b)

(c)

Fig. 10—Profile fitting of XRD pattern by Rietveld refinement for precipitates from Grade100 steel, electrolytically dissolved using 10 pct AAdissolution. (a) Overall XRD pattern profile fitting, (b) calculated diffraction pattern for Ti0.9Nb0.1N, (c) calculated diffraction pattern forTi0.77Nb0.23C0.5N0.5, (d) calculated diffraction pattern for Ti0.5Nb0.5C0.5N0.5, (e) calculated diffraction pattern for Nb0.7Ti0.3C0.5N0.5, and (f) calcu-lated diffraction pattern for Nb0.48Mo0.28Ti0.21V0.03C. Obs: observed diffraction pattern; Calc: calculated diffraction pattern; Diff: differencebetween the calculated and observed pattern.


Table V. Bragg R-Index (RBragg) and Lattice Parameters for Different Phases in Grade100 Steel by Two Dissolution Methods

Phases

Electrolytic (10 pct AA) Chemical (HCl)

RBragg Lattice Parameters (nm) RBragg Lattice Parameters (nm)

Ti0.9Nb0.1N 0.568 0.4248 2.145 0.4250Ti0.77Nb0.23C0.5N0.5 0.426 0.4282 1.062 0.4282Ti0.5Nb0.5C0.5N0.5 0.085 0.4329 0.173 0.4329Nb0.7Ti0.3C0.5N0.5 0.107 0.4418 0.580 0.4419Nb0.48Mo0.28Ti0.21V0.03C 0.078 0.4392SiO2 N/A N/A 0.108 0.4622

908580757065605550454035

9,0008,0007,0006,0005,0004,0003,0002,0001,000

0-1,000-2,000

908580757065605550454035

9,0008,0007,0006,0005,0004,0003,0002,0001,000

0-1,000-2,000

908580757065605550454035

9,0008,0007,0006,0005,0004,0003,0002,0001,000

0-1,000-2,000

Nb0.7Ti0.3C0.5N0.5

Ti0.5Nb0.5C0.5N0.5

Nb0.48Mo0.28Ti0.21V0.03C

(d)

(e)

(f)

Fig. 10—Continued.


results in formation of an (Fe, M)xCyNz compound.[4]

Additional study is needed to understand whetherthe presence of Fe in the EDX spectra in this work(e.g., Figure 4) is from the steel matrix or from an(Fe,M)xCyNz compound.

IV. DISCUSSION

A. Determination of the Relative Abundance of DifferentPhases via Rietveld Refinement

The relative abundance (expressed as wt pct) of theconstituent crystalline phases in the extracted residue

was determined from Eq. [2] by Rietveld refinement andis shown in Table VIII (electrolytic dissolution) andTable IX (chemical dissolution). The total wt pct ofprecipitates relative to the amount of steel is also shownin these tables, as is the amount of each precipitaterelative to the steel.It is clear that nanoprecipitates with an average

composition of Nb0.48Mo0.28Ti0.21V0.03C (£10 nm) arethe majority crystalline phase in both chemically andelectrolytically extracted residues. The relative amountsdiffered by approximately 10 pct, i.e., 0.140 wt pct(electrolytic dissolution) and 0.157 wt pct (chemicaldissolution), respectively. However, there is relativelypoor agreement for the other types of precipitates with

908580757065605550454035302520

4,000

3,500

3,000

2,500

2,000

1,500

1,000

500

0

-500

-1,000

Obs

Diff

Calc

908580757065605550454035302520

4,000

3,500

3,000

2,500

2,000

1,500

1,000

500

0

-500

-1,000

Ti0.9Nb0.1N

908580757065605550454035302520

4,000

3,500

3,000

2,500

2,000

1,500

1,000

500

0

-500

-1,000

Ti0.77Nb0.23C0.5N0.5

(a)

(b)

(c)

Fig. 11—Profile fitting of XRD pattern by Rietveld refinement for residue from Grade100 steel, chemically dissolved using HCl dissolution.(a) Overall XRD pattern profile fitting, (b) calculated diffraction pattern for Ti0.9Nb0.1N, (c) calculated diffraction pattern forTi0.77Nb0.23C0.5N0.5, (d) calculated diffraction pattern for Ti0.5Nb0.5C0.5N0.5, (e) calculated diffraction pattern for Nb0.7Ti0.3C0.5N0.5, and (f) calcu-lated diffraction pattern for amorphous SiO2 and nanoprecipitates. Obs: observed diffraction pattern; Calc: calculated diffraction pattern; Diff:difference between the calculated and observed pattern.


the two extraction methods. There are two possiblereasons for the discrepancy. First, there is a largeamount of amorphous SiO2 in the chemically extractedresidue, which could affect those results. Second, asmaller amount of precipitates produced by electrolyticdissolution was used for XRD analysis (approximately10 pct of the mass used for chemical dissolution)because of the inefficiency in precipitate extraction.Therefore, errors could be amplified, especially for the

minor precipitate phases. Overall, if the same amountsof residues are collected by both chemical and electro-lytic dissolution methods, then the latter method shouldbe more reliable because determination of the relativeabundance of nanoprecipitates is not hindered by SiO2.However, electrolytic dissolution is much slower than

Table VII. C and N Content Calculations in PrecipitatesUsing Two Dissolution Methods

DissolutionMethod Electrolytic (10 pct AA) Chemical (HCl)

Precipitate Ti0.96Nb0.04N Ti0.95Nb0.05NTi0.77Nb0.23C0.07N0.93 Ti0.77Nb0.23C0.07N0.93

Ti0.5Nb0.5C0.14N0.86 Ti0.5Nb0.5C0.14N0.86

Nb0.7Ti0.3C0.89N0.11 Nb0.7Ti0.3C0.90N0.10

908580757065605550454035302520

4,000

3,500

3,000

2,500

2,000

1,500

1,000

500

0

-500

-1,000

Ti0.5Nb0.5C0.5N0.5

908580757065605550454035302520

4,000

3,500

3,000

2,500

2,000

1,500

1,000

500

0

-500

-1,000

Nb0.7Ti0.3C0.5N0.5

908580757065605550454035302520

4,000

3,500

3,000

2,500

2,000

1,500

1,000

500

0

-500

-1,000

Amorphous SiO2 and nano-precipitates

(d)

(e)

(f)

Fig. 11—Continued.

Table VI. Lattice Parameters for Binary Carbidesand Nitrides

LatticeParameter (nm) TiN TiC NbN NbC

a 0.42417 0.43274 0.43927 0.44698


chemical dissolution. In any case, our main concern isthe nanoprecipitates as these are the major contributorto precipitate strengthening. The agreement, as men-tioned above, using the two dissolution methods is quitegood.

Another way to determine the amount of nanopre-cipitates (£10 nm) is based on the amount of Mo. ICPanalysis of the supernatant indicates that most of theMo is present in solid solution in the ferrite. Theremaining Mo, presented in Table III, is only present inthe nanoprecipitates (£10 nm), as confirmed by TEMEDX analysis (Figures 4, 8, and 9). Therefore, theamount of nanoprecipitates (£10 nm) can be determinedusing Eq. [3]. This is based on the chemistry of thenanoprecipitates and the total amount of Mo availablefor precipitation.

wtMo

AMo¼ wtppt

Appt� at: fractMo ½3�

where wtMo is the total amount (weight) of Mo availablefor precipitation, which was determined from the totalamount of Mo in steel minus that in the supernatant(ICP analysis); wtppt is the amount (weight) of nano-precipitate (£10 nm) to be determined; AMo is theatomic weight of Mo; Appt is the molecular weight ofthe nanoprecipitate (£10 nm); at. fractMo is the atomicfraction of Mo in the nanoprecipitate (£10 nm), fromthe nanoprecipitate composition determined fromTEM-EDX microanalysis, i.e., 0.28.

Figure 12 shows a comparison of the amount ofnanoprecipitates based on Rietveld refinement (viachemical and electrolytic dissolution) and that basedon the Mo amount. The amounts of nanoprecipitates(£10 nm) determined by the three methods are similar.

B. Volume Fraction and Number Densityof Nanoprecipitates

The volume fraction and number density of thenanoprecipitates (£10 nm) can be determined from their

weight fraction, which has been explained in theprevious section. The nanoprecipitates (£10 nm) havea NaCl-type structure. From the precipitate chemistry,the density of the precipitates can be expressed math-ematically as follows[30]:

qppt ¼nAppt

VNA½4�

where qppt is the density of the nanoprecipitate, n is thenumber of formula units in the unit cell (n = 4), Appt isthe molecular weight of the precipitate, V is the volumeof the unit cell, and NA is Avogadro’s number(6.023 9 1023 atoms/mol).The molecular weight Appt can be calculated in the

following equation, using Nb0.48Mo0.28Ti0.21V0.03C asan example.

Appt ¼ANb � 0:48þ AMo � 0:28þ ATi � 0:21

þ AV � 0:03þ AC ½5�

where Appt is the molecular weight of the precipitate,ANb is the atomic weight of Nb, AMo is the atomic

Table IX. Relative Abundance of Constituent Crystalline Phases Using Chemical Dissolution

PhasesRelative Abundance (wt pct)of Precipitates in Residue

Experimental ExtractionYield (wt pct)

Fraction of Precipitates Relativeto Mass of Steel Dissolved

Ti0.96Nb0.05N 1.6 pct 0.792 pct 1.3 9 10�4

Ti0.77Nb0.23C0.07N0.93 3.4 pct 2.7 9 10�4

Ti0.5Nb0.5C0.14N0.86 0.8 pct 0.6 9 10�4

Nb0.7Ti0.3C0.90N0.10 3.2 pct 2.5 9 10�4

Nb0.48Mo0.28Ti0.21V0.03C 19.8 pct 1.57 9 10�4

Table VIII. Relative Abundance of Constituent Crystalline Phases Using Electrolytic Dissolution

PhasesRelative Abundance (wt pct)of Precipitates in Residue

Experimental ExtractionYield (wt pct)

Fraction of Precipitates Relativeto Mass of Steel Dissolved

Ti0.96Nb0.04N 10.8 pct 0.361 pct 3.8 9 10�4

Ti0.77Nb0.23C0.07N0.93 26.4 pct 9.6 9 10�4

Ti0.5Nb0.5C0.14N0.86 8.5 pct 3.1 9 10�4

Nb0.7Ti0.3C0.89N0.11 15.5 pct 5.6 9 10�4

Nb0.48Mo0.28Ti0.21V0.03C 38.8 pct 1.4 9 10�3

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

Wt%

of

nan

o-p

reci

pit

ates

Grade100 steel

Rietveld refinement via chemical dissolution

Mo amount from steel chemistry and ICP analysis of HCl supernatant

Rietveld refinement via electrolytic dissolution

Fig. 12—The amount of nanoprecipitates (£10 nm) compared withthe original mass of the steel.


weight of Mo, ATi is the atomic weight of Ti; AV is theatomic weight of V, and AC is the atomic weight of C.

The volume fraction of the nanoprecipitates (£10 nm)can be calculated as

vol fractionppt ¼wt fractionppt � qFe

qppt

½6�

where vol fractionppt is the volume fraction of the nano-precipitates in one unit volume of steel, qFe is the densityof Fe; qppt is the density of nanoprecipitates, and wtfractionppt is the weight fraction of the nanoprecipitates.

From the preceding TEM precipitate characterization,the nanoprecipitate diameter with the largest numberdistribution is approximately 4.5 nm. Assuming all thenanoprecipitates (£10 nm) are spherical and using the4.5 nm diameter, the number density of the nanoprecip-itates in one unit volume of steel can be calculated basedon the total volume of the nanoprecipitates and thevolume of a single precipitate, which can be expressed as

#density ¼ volpptvolsingle

½7�

volsingle ¼4pr3

3½8�

where volppt is the total volume of nanoprecipitates in 1lm3 of steel, volsingle is the volume of a single nanopre-cipitate; and r is the radius of a nanoprecipitate(£10 nm) with the largest number distribution deter-mined by DF imaging.

The weight fraction, volume fraction and numberdensity of the nanoprecipitates (mainly £10 nm) inGrade100 steel using the two dissolution methods areshown in Table X. If the precipitation-strengtheningcontribution is calculated using Eq. 1, the contributionis in the 185 to 195 MPa range (using the volumepercents of 0.140 pct and 0.157 pct), which represents 27to 28 pct of the strengthening in the Grade100 steel.[11]

C. Mass balance Comparison between Rietveldand ICP Analysis

There are two ways to calculate the amounts of Nb,Ti, Mo, and V that are present in the precipitates, i.e.,through Rietveld refinement and ICP analysis of thesupernatant. From Rietveld refinement, the relativeabundance of different phases in the precipitates isobtained. Based on precipitate chemistries, the totalamount of microalloying elements present in theprecipitates can be calculated. From an ICP analysisof the supernatant (Table III), the total amount ofmicroalloying elements dissolved in the ferrite matrixcan be obtained. The amount of alloying elementspresent in the precipitates can be determined indirectlyby subtracting the amount of alloying elements in thesupernatant from the total in the steel. A comparisoncan then be made between the two methods as shownin Table XI. The agreement generally is good, withthe largest difference occurring for Nb. The errorsare largest for Rietveld analysis and include errorsarising from EDX analysis and XRD pattern profilefitting.

V. CONCLUSIONS

Matrix dissolution techniques have been developed toextract and quantify precipitates from a Grade100microalloyed steel. The following conclusions can bedrawn from this investigation:

1. Matrix dissolution is a promising technique toquantify precipitates in microalloyed steels, providing

Table X. Weight Fraction, Volume Fraction, and Number

Density of Nanoprecipitates

PrecipitateElectrolytic(10 pct AA)

Chemical(HCl)

Weight fraction (wt pct) 0.140 0.157Volume fraction (vol pct) 0.149 0.166Diameter with the largestnumber distribution (nm)

4.5 4.5

Number density (# per lm3) 31145 34770

Table XI. Mass Balance–Weight Percent of Elements in Steel and in Precipitates for Grade100 Steel

Wt pct Nb total wt pct of Nb in steel 0.094total wt pct of Nb in precipitates (Rietveld refinement) 0.104total wt pct of Nb in precipitates (ICP analysis) 0.091difference* –14.0 pct

Wt pct Ti total wt pct of Ti in steel 0.06total wt pct of Ti in precipitates, (Rietveld refinement) 0.045total wt pct of Ti in precipitate (ICP analysis) 0.046difference* 1.6 pct

Wt pct Mo total wt pct of Mo in steel 0.301total wt pct of Mo in precipitate (Rietveld refinement) 0.0439total wt pct of Mo in precipitate (ICP analysis) 0.0437difference* 0.5 pct

Wt pct V total wt pct of V in steel 0.047total wt pct of V in precipitate (Rietveld refinement) 0.003total wt pct of V in precipitate (ICP analysis) 0.002difference* –7.7 pct

*Difference: (amount from ICP minus amount from Rietveld refinement) divided by amount from ICP analysis.


quantitative information based on samples that arerepresentative of the steel.

2. Different sized precipitates can be extracted success-fully using chemical (HCl) and electrolytic dissolu-tion (10 pct AA, composed of 10 pct acetylacetone,1 pct tetramethylammonium chloride, and metha-nol). Chemical dissolution is more efficient forextraction; however, SiO2 comprises a large portionof the residue from chemical dissolution.

3. Matrix dissolution and carbon replicas yieldconsistent results for precipitate size and chemistry.Carbon replicas validate the matrix dissolutiontechnique.

4. Based on TEM imaging and EDX microanalysis,five groups of precipitates were identified in theGrade100 microalloyed steel in terms of size andcomposition. Nanoprecipitates (£10 nm) with anaverage chemistry corresponding to Nb0.48Mo0.28-Ti0.21V0.03C are the most prevalent crystalline phasein the extracted residues using both chemical andelectrolytic dissolution methods.

5. Rietveld refinement of XRD patterns can be usedsuccessfully to identify and determine the relativeamounts of different precipitate phases, making itpossible to determine the volume fraction of nano-precipitates in microalloyed steels.

6. Mass balance of Nb, Ti, Mo, and V is consistentusing Rietveld refinement and ICP analysis.

ACKNOWLEDGMENTS

The authors thank the Natural Sciences and Engi-neering Research Council (NSERC) of Canada andEvraz Inc., NA for financial support. Provision ofexperimental materials by Evraz Inc. NA is acknowl-edged as well.

REFERENCES1. D. Bai, M.A. Cooke, J. Asante, and J. Dorricott: Patent US

6,682,613 B2, 2004.2. S. Wolf: JOM, 1967, vol. 19, pp. 22–28.

3. T. Gladman: The Physical Metallurgy of Microalloyed Steels, TheInstitute of Materials, London, UK, 1997.

4. K. Poorhaydari-Anaraki: Ph.D. Dissertation, University of Al-berta, Alberta, Canada, 2005.

5. A.J. DeArdo: Int. Mater. Rev., 2003, vol. 48 (6), pp. 371–402.6. S. Akhlaghi and D.G. Ivey: Can. Metall. Q., 2002, vol. 41 (1),

pp. 111–19.7. R. Lagneborg, T. Siwecki, S. Zajac, and B. Hutchinson: Scand.

J. Metall., 1999, vol. 28 (5), pp. 186–241.8. U. Lagerpusch, B. Anczykowski, and E. Nembach: Phil. Mag. A,

2001, vol. 81 (11), pp. 2613–28.9. M.G. Burke and M.K. Miller: J. Electron Microsc. Tech., 1988,

vol. 8, pp. 201–10.10. K. Stiller and M. Hattestrand: Microsc. Microanal., 2004, vol. 10,

pp. 341–48.11. J. Lu: Ph.D. Dissertation, University of Alberta, Alberta, Canada,

2009.12. J. Chakraborty, D. Bhattacharjee, and I. Manna: Scripta Mater.,

2008, vol. 59, pp. 247–50.13. J. Chakraborty, D. Bhattacharjee, and I. Manna: Scripta Mater.,

2009, vol. 61, pp. 604–07.14. W. Huang: Metall. Trans. A, 1991, vol. 22A, pp. 1911–20.15. L.E. Collins, M.J. Godden, and J.D. Boyd: Can. Metall. Q., 1983,

vol. 22 (2), pp. 169–79.16. V.B. Ginzburg: Metallurgical Design of Flat Rolled Steels, Marcel

Dekker Inc, New York, NY, 2005.17. F.G. Wilson and T. Gladman: Int. Mater. Rev., 1988, vol. 33 (5),

pp. 211–86.18. S. Dawson, N.D.G. Mountford, I.D. Sommerville, and A.

McLean: Ironmaking Steelmaking, 1988, vol. 15 (10), pp. 54–55.19. ASTM E194-90, ‘‘Standard Test Method for Acid-Insolute Con-

tent of Copper and Iron Powders’’, 1990, pp. 1–2.20. ‘‘Physical Constants of Inorganic Compounds’’, CRC Handbook

of Chemistry and Physics, Knovel, Binghamton, NY, 2004.21. Y. Yoshida, H. Yoshiko, I. Kanji, and K. Yoshikazu: Kawasaki

Steel Giho, 1980, vol. 12 (4), pp. 653–64.22. H.M. Rietveld: Acta Crystallogr., 1967, vol. 22, pp. 151–52.23. H.M. Rietveld: J. Appl. Crystallogr., 1969, vol. 2, pp. 65–71.24. R.J. Hill and C.J. Howard: J. Appl. Crystallogr., 1987, vol. 20,

pp. 467–74.25. S.G. Hong, K.B. Kang, and C.G. Park: Scripta Mater., 2002,

vol. 46 (2), pp. 163–68.26. JCPDS: International Center for Diffraction Data, Newtown

Square, PA, 1996, PDF#00-038-1420, PDF#00-032-1383, PDF#00-038-1155, PDF#00-039-1425, and PDF#00-038-1364.

27. E.L. Brown and A.J. DeArdo: Proc. Int. Conf. on the Thermo-mechanical Processing of Microalloyed Austenite. A.J. DeArdo,G.A. Ratz, and P.J. Wray, eds., AIME, Warrendale, PA, 1982,pp. 319–41.

28. S. Kanazawa, A. Nakashima, K. Okamoto, K. Tanabe, and S.Nakazawa: Trans. Jpn. Inst. Met., 1967, vol. 8 (2), pp. 113–19.

29. W.R. Bandi: Science, 1977, vol. 196 (4286), pp. 136–42.30. W.D. Callister: Materials Science and Engineering: An Introduc-

tion, 5th ed., Wiley, New York, NY, 2000.


Matrix Dissolution Techniques Applied to Extract and ... · Matrix Dissolution Techniques Applied...

Documents

Transcript of Matrix Dissolution Techniques Applied to Extract and ... · Matrix Dissolution Techniques Applied...