Elucidation of the role of hexamine and other precursors in the formation

of magnetite nanorods and their stoichiometry

Hema Singh,a Shrikant Bhagwat,a Samuel Jouen,b Benoıt Lefez,b

Anjali A. Athawale,*aBeatrice Hannoyer

band Satishchandra Ogale*

c

Received 24th August 2009, Accepted 19th January 2010

First published as an Advance Article on the web 19th February 2010

DOI: 10.1039/b917407a

Hexamine is known to assist anisotropic growth of metal oxides and the same is also found

to be true for magnetite nanosynthesis. In this work we elucidate the role of hexamine and other

precursors in the formation of magnetite nanorods by the hydrothermal route and their

stoichiometry. Various others hydrolyzing agents such as sodium hydroxide (NaOH), sodium

hydroxide + hexamine, ammonia (NH3), ammonia + formaldehyde are also studied. The

synthesized nanoparticles are characterized with the help of various techniques such as X-ray

diffraction (XRD), FT-IR spectroscopy, UV-VIS-NIR spectroscopy, transmission electron

microscopy (TEM), Mossbauer spectroscopy and SQUID magnetization measurements. It is

found that only when ferric chloride, ferrous ammonium sulfate (FAS) and hexamine are used,

well defined nanorods are formed. When sodium hydroxide and hexamine are used as a

hydrolyzing system nearly spherical nanoparticles with small size (B13 nm) are formed, as

compared to the case of sodium hydroxide alone which leads to bigger cube like nanoparticles.

Interestingly the decomposition products of hexamine do not lead to nanorod formation. Thus,

slow decomposition of hexamine at elevated temperature and the consequent slow rise in pH is

the key to the anisotropic growth of the iron oxide system.

Introduction

Low dimensional (quasi 1D, 2D nanomaterials such as nano-

wires, nanotubes, nanobelts and nanoribbons) have been the

focus of considerable interest because of their fundamental

importance and potential applications in nanoscale device

systems with novel electrical, magnetic and optical properties.1

Iron oxide is one of the most important magnetic materials

(with a projected 100% spin polarization) and has numerous

potential applications, for example, in ultrahigh density

magnetic storage devices,2,3 drug delivery,4 tissue-repair

engineering5 etc. It is also a bio-friendly material. Several

methods have been reported for the synthesis of spherical iron

oxide nanoparticles, which have attracted significant research

attention.6–8

Early studies had reported the formation of micro-

metre-sized fibrous structures by thermally oxidizing iron.9

Recently, iron oxide microfibres and nanowires have also been

obtained by oxidizing iron in pure oxygen at about 700 1C.10

Fu et al.11 have achieved considerable success in the creation

of iron oxide nanorods by using a mixture of CO2, SO2 and

NO2 gases together with a small amount of H2O vapor to react

with iron atB550 1C. Fe3O4 nanorods with average diameters

of 40–50 nm and length B1 mm have also been synthesized

through hydrolysis of FeCl3 and FeSO4 solutions containing

urea at the temperature of B95 1C in reflux conditions for

12 h.12 The hydrothermal technique has also been widely used

for the growth of inorganic crystals because it is a less

polluting low temperature technique, and leads to anisotropic

crystal growth in solution. A number of papers dealing with

iron oxide formation under hydrothermal conditions have

been published.13–17 Uniform single-crystal Fe3O4 nanorods

with an average diameter of 25 nm and length of 200 nm have

been synthesized via a soft-template-assisted hydrothermal

route at 120 1C for 20 h using ethylene diamine not only as

a base source but as a soft template and benzene as an oil

membrane.16 Fe3O4 nanowires with a narrow diameter

distribution centered at 15 nm and length up to several

microns have been synthesized by a simple hydrothermal route

with the assistance of polyethylene glygol (PEG) 400.17

Hexamine (hexamethylene tetra amine) is a highly water

soluble, non-ionic tertiary amine derivative and results indicate

that it acts as a shape inducing molecule. It has been used for

the synthesis of anisotropic alumina, zirconia, ceria, zinc oxide

etc.18,19 It is reported that hexamine acts as a directing agent

for the synthesis of zinc oxide nanorods where zinc oxide

nanoparticles are used as seeds for the growth.20 Recently, we

have also utilized hexamine for the synthesis of magnetite

nanorods.21 Unfortunately an understanding of the precise

role of hexamine in these growth processes is still lacking. Here

we attempt to elucidate the role of hexamine though a series of

careful experiments in which different hydrolyzing agent

systems are used and the implications for the Fe(II) precursor

behavior are examined. Variations of experimental conditions

aDepartment of Chemistry, University of Pune, Pune 411 007, IndiabGroupe de Physique des Materiaux, Universite de Rouen, UMR 6634CNRS B.P. -12 St Etienne du Rouvray CEDEX, France

cNational Chemical Laboratory, Dr Homi Bhabha Road, Pashan,Pune, India. E-mail: sb.ogale@ncl.res.in; Fax: +91-20-2590-2636;Tel: +91-20-2590-2260

3246 | Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 This journal is �c the Owner Societies 2010

PAPER www.rsc.org/pccp | Physical Chemistry Chemical Physics

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online / Journal Homepage / Table of Contents for this issue

such as precursors, concentration of the precursors and

temperature result in significant changes in morphology

and thereby the properties. These changes are tracked and

analyzed in this paper using a variety of techniques.

Experimental

In a typical procedure, high purity ferric chloride (FeCl3)

(2 mM), ferrous ammonium sulfate (FAS) (1mM) and hexamine

(1 M), all from Qualigen chemicals, were dissolved in 50 ml of

double distilled water. The mixture was then loaded in a

stainless steel (SS) autoclave with Teflon cup. The SS reactor

was then heated to a temperature between 150–170 1C at a

pressure of B50 lb in�2 (3.45 Bar) for three hours in a heating

mantle. The autoclave was then cooled to room temperature

and the contents of the autoclave were washed with double-

distilled water in a centrifuge (B5000 rpm) till the pH of the

supernatant solution was B8.0–8.5. The precipitate was then

dried in an oven at B80 1C for 24 h until constant weight was

obtained. We identify this sample with a label FH1. The same

basic procedure was also adopted for synthesizing nano-

particles by using ferrous sulfate as Fe2+ source instead of

ferrous ammonium sulfate (sample labeled as FH4). Further

experiments with other combinations of hydrolyzing agents

were also performed. These were guided by literature reports22

that in acidic media hexamine undergoes slow thermal

decomposition into formaldehyde and ammonia, and the

consequent slow rise in pH is an important factor that

contributes to the anisotropy of the nanorods. It was important

to analyze therefore as to whether slow decomposition of

hexamine, presence of ammonia or combination of ammonia

and formaldehyde is responsible for the formation of rods.

The agents used and the corresponding sample labels are

summarized in Table 1.

The nanopowder samples were characterized by X-ray

diffraction (XRD), Fourier transform infra red spectroscopy

(FTIR), Transmission electron microscopy (TEM), UV-Vis-NIR

diffuse reflectance spectroscopy and SQUID magnetometry.

UV-Vis spectroscopy measurements of the solid samples at

room temperature were carried out on a Jasco UV-Vis spectro-

photometer (V570 UV-VIS-NIR) at a resolution of 1 nm. and

a Perkin Elmer Lambda 9 spectrophotometer equipped with

an integrating sphere over the 200–2500 nm wave-length

range. The FTIR spectra were recorded on Schimadzu double

beam spectrophotometer CFI-IR 8400. The dry powders were

ground with spectroscopic grade KBr powder. KBr was used

as a reference material. The weight ratio of KBr to sample was

maintained to be 100 : 1. The spectra of the samples were

recorded from 400 to 4000 cm�1. The structural phase analysis

of the samples was carried out using Philips PW 1830 X-ray

diffractometer with Ni filtered Cu-Ka radiation of wavelength

1.5405 A. The TEMmeasurements were performed on a JEOL

JEM-1200EX instrument operating at 120 kV, camera length

of 80 cm and field limited aperture of 100 mm. Prior to TEM

measurements, the samples were dispersed in methanol and a

drop of the solution was poured on carbon-coated copper grid

of 400 mesh size. The film formed on the TEM grids was

allowed to dry for two minutes following which the extra

solvent was removed using a blotting paper. The image and

diffraction patterns were obtained at an accelerating voltage of

120 kV. Mossbauer measurements were performed in trans-

mission geometry between 30 and 300 K using a constant

acceleration spectrometer with a 57Co(Rh) source. Computer

fitting of the spectra was performed to obtain the hyperfine

interaction parameters and field distributions; the isomer shift

(d) values were referred to metallic iron foil. Raman spectra of

the samples were measured at room temperature on a mRaman

spectrometer (Aramis, Horiba Jobin Yvon, France), in

confocal configuration, using a He–Ne laser (632.8 nm) as

an excitation source at a very low power to avoid degradation

of the sample. The magnetization measurements were performed

using the Quantum Design MPMSSquid system.

Results and discussion

The experiments performed and the corresponding shape

outcomes are summarized in Table 1.

Case 1: sample FH1

As stated earlier, this synthesis was carried out using hexamine

which acts as a directing agent as well as hydrolyzing agent.21

NaOH was not present in this case. The XRD of the synthe-

sized sample is shown in Fig. 1 (FH1). The peaks appearing at

‘d’ values 2.96, 2.53, 2.09, 1.71, 1.61 A correspond to the

magnetite phase as per the reported data (JCPDS 85-1436). It

should however be recalled 21 that magnetite and maghemite

XRD signatures are too close hence other techniques are

needed to ensure the presence of one of these two phases.

The as-recorded XRD pattern shows spinel peaks riding on

background humps, suggesting two different diffracting length

scales characteristic of anisotropic growth.

Formation of nanorods was confirmed by TEM analysis of

the sample. Transmission electron micrographs of Fig. 2

(FH1) clearly show the rod formation. The inset shows the

edge structure of a nanorod. The selected area electron

diffraction pattern depicted as an inset in Fig. 2 (FH1) shows

Table 1 Description of samples studied

Sample code no. Ferric ion source Ferrous ion source Hydrolyzing agent used Shape of the particle

FH1 2 mM FeCl3 1 mM FAS 1 M hexamine RodsFH2 2 mM FeCl3 1 mM FAS 1.5 M NaOH CubeFH3 2 mM FeCl3 1 mM FAS 1 M hexamine+ 1.5 M NaOH SphericalFH4 2 mM FeCl3 1 mM FeSO4 1 M hexamine Rods + sphericalFH5 2 mM FeCl3 1 mM FAS NH3 FacetedFH6 2 mM FeCl3 1 mM FAS NH3 + HCHO Spherical

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 | 3247

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online

the crystalline nature of the sample, revealed by bright spots

on the rings. To gain a better understanding of the growth

process of magnetite nanorods infrared spectroscopic

measurement and analysis were carried out as a function of

processing duration and these data are given in Fig. 3(A).

Fig. 3(B) shows the data for the significant region on an

expanded scale for clarity.

Fig. 3A (curve a) illustrates the IR spectrum taken after half

an hour of synthesis.

Important diagnostic bands are seen at nmax/cm�1 896, 796

andB628 in addition to a band at nmax/cm�1 B3138. The first

three together correspond to goethite signatures while the last

one is due to adsorbed water. The bands at nmax/cm�1 B 896

(d-OH) and 796 (g-OH) are the OH bending bands of the (001)

plane of goethite.23 The band at nmax/cm�1 B 628 corresponds

to symmetric Fe–O stretching vibration which lies in the (010)

plane. It is thus clear that in the initial stage of the reaction

goethite is formed. Further, it is also seen to persist in the

FTIR spectrum for one hour growth (Fig. 3A curve b) almost

at the same position but with slightly reduced intensity. After

three hours (Fig. 3A curve c) the bending vibration bands at

nmax/cm�1 B 896 (d-OH) and 796 (g-OH) attributed to

goethite are found to decrease significantly in comparison

to those in the FTIR spectrum for half or one hour

processing cases.

In addition to this, a band at nmax/cm�1 B 580 is seen to

emerge (Fig. 3B curve c with arrows marked M for magnetite

and G for goethite) which clearly reveals the formation of

magnetite form of iron oxide, though the characteristic band

area is not fully separated for the two phases.24,25

Case 2: system FH2

The XRD pattern for material formed when FeCl3 and FAS

are used as Fe3+ and Fe2+ sources and only NaOH as a

hydrolyzing agent (without hexamine) is also depicted in Fig. 1

(FH2). The peaks at ‘d’ values 2.96, 2.52, 2.09, 1.71, 1.61 and

1.57 A in the XRD pattern once again show the formation

of magnetite nanoparticles, although as stated earlier the

questions about stoichiometry need to be settled through other

techniques as discussed later. The size of the nanoparticles

Fig. 1 X-Ray diffraction data for various cases examined. The

sample numbers correspond to the cases given in Table 1.

Fig. 2 TEM data for various samples described in Table 1. Insets

show electron diffraction patterns.

Fig. 3 (A) FTIR data revealing kinetics of growth. Curves a, b and c

correspond to 12Hr, 1 Hr and 3 Hr processing. (B) shows data on

expanded scale.

3248 | Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 This journal is �c the Owner Societies 2010

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online

is found to be B30 nm when calculated using Scherrer’s

equation.

The TEM micrograph presented in Fig. 2 (FH2) shows

cube-like morphology that is completely different as compared

to the case of FH1, wherein hexamine was used in place of

NaOH. The particle size is B30 nm, consistent with the XRD.

The selected area electron diffraction (SAED) pattern of the

sample reveals its crystalline nature (bright dots on the ring

pattern). In the FTIR spectrum for this sample (Fig. 4 (FH2))

a broad band at nmax/cm�1 B 580 is seen which can be

assigned to magnetite form of iron oxide. The G, M and H

symbols in Fig. 4 represent goethite, magnetite and hexamine,

respectively.

Case 3: system FH3

When hexamine was added along with NaOH (FH3), the

positions of the XRD peaks were the same as in the case of

NaOH alone (Fig. 1, FH2), but the peaks appeared to

be broader, with particle sizes in the range of 10 � 2 nm.

This showed that nanoparticles formed by using the system

of NaOH and hexamine are of considerably smaller size

than those formed when NaOH alone is used as hydrolyzing

agent.

The TEM micrograph for this case depicted in Fig. 2 (FH3)

shows mixed morphology but with a fairly uniform particle

size. The particle size appears to be smaller (B13 nm) than

that for the FH2 case. This confirms that Hexamine acts as a

stabilizer of particle size via selective deceleration of growth

rate and prevention of particles agglomeration.

In the FTIR spectrum shown in Fig. 4 (FH3), apart from

the absorption band nmax/cm�1 B 578 corresponding to

magnetite, few more peaks are seen at nmax/cm�1 661, 848,

1009, 1111 (marked as H0). These closely correspond to the

peaks marked H on the hexamine curve, albeit with a small

shift. This implies that the sample FH3 has hexamine capping.

It is further interesting to point out that 1009 cm�1 peak

represents the stretching vibrations of C–N bonds in tertiary

amines present in hexamine. Thus the capping of hexamine is

in the form of the cage molecule itself. When only hexamine

was used however (Fig. 3, curve c), this 1009 cm�1 peak is

totally absent indicating that hexamine has undergone thermal

decomposition into formaldehyde and ammonia in acidic

medium.

Case 4: system FH4

In this experiment Fe2+ source was changed to FeSO4 while

keeping the Fe3+ source as FeCl3 with only hexamine as

hydrolyzing agent similar to the FH1 case. Once again the

XRD pattern (Fig. 1 (FH4)) confirms the spinel structure of

the material which could correspond to magnetite, maghemite

or the mixture of the two. Interestingly, the morphology is

found to be a mixture of spherical nanoparticles and nanorods

in almost comparable proportion as shown in Fig. 2 (FH4), in

contrast to FH1 wherein primarily nanorods are observed.

The FTIR spectrum (Fig. 4 (FH4)) of the sample shows peak

at nmax/cm�1 B 580 corresponding to magnetite phase along

with some impurity of goethite.

UV-Vis-NIR diffuse reflectance: FH1–FH4

As stated earlier, in order to confirm whether the material is

magnetite or maghemite it is essential to perform other

measurements, since XRD patterns as well as infrared spectra

for the two phases magnetite and maghemite are too close. In

this context, we performed optical measurements to get

electronic spectra. Fig. 5 presents the UV-visible-near infrared

absorption spectra of four samples synthesized under different

experimental conditions, along with some data for two

reference compounds magnetite and maghemite.

The absorption band at 7000 cm�1 and the rise in its

absorption level between 4000 and 25 000 cm�1 represent

an inter-valence charge transfer Fe2+ to Fe3+ (IVCT) as

in magnetite (reference compound), and this electronic

delocalization requires a ferromagnetic coupling between the

cations Fe2+ and Fe3+ ions.26,27 Although some differences

are noted in the data for the four cases of interest, it is hard to

make a precise statement about the stoichiometry because the

appearance of a band towards 7000 cm�1 and the rise in the

absorption level between 4000 and 25 000 cm�1 has a

non-linear dependence on the rising Fe2+ contribution. For

this purpose we resorted to Mossbauer spectroscopy analysis

discuss later.

Magnetization: FH1–FH4

The results of magnetization measurements (room temperature

hysteresis loops) on FH1 to FH4 are shown in Fig. 6. The

values of magnetization at 3000 Oe, which is close to but not

Fig. 4 FTIR data for the cases of FH2, FH3 and FH4. Data for

hexamine is also shown.

Fig. 5 UV-visible-near infrared absorption spectra for FH1–FH4,

along with data for magnetite and maghemite.

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 | 3249

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online

strictly the saturation magnetization, for FH1, FH2, FH3 and

FH4 are B52, 81, 60 and 48 emu g�1, respectively.

Note that the magnetization is highest for the FH2 case

which the optical data suggests to be not far from magnetite

stoichiometry. The magnetization is also fairly high in the FH3

case, but considerably lower in the FH1 and FH4 cases. The

values of coercive field (remanence) for FH1, FH2, FH3

and FH4 are 48(6.8), 61(6.6), 0(0) and 84(10) Oe(emu g�1),

respectively. If we represent Fe3�xO4 as the basic phase, then x

appears to vary in different samples, which is consistent with

the optical spectroscopy results discussed earlier. This also

implies changes in Fe3+ to Fe2+ ratio, which can be verified

by Mossbauer spectroscopy, discussed next.

Mossbauer spectroscopy: FH1–FH4 (with FH5 and FH6)

The Mossbauer measurements of FH1 to FH6 samples are

depicted in Fig. 7.

These spectra are described with two or three components,

one for Fe3�xO4 and its two sextets (underlined in grey), one

for goethite as a rather wide sextet with hyperfine field

distribution, and the third doublet assigned to particles of

iron oxide in the superparamagnetic state. The Mossbauer

spectra of samples FH5 and FH6, shown here for comparison,

are characteristic of single phase samples. They are described

with a hyperfine field distribution, showing relaxation effects

in agreement with their smaller particulate size and a large

contribution of surface disorder and inter-particle inter-

actions. The hyperfine interaction parameters and contribution

of the different spectral components are summarized in

Table 2.

The Mossbauer spectrum of magnetite at room temperature

is commonly interpreted as a superposition of two patterns,

one due to trivalent iron on tetrahedral A sites and the other

to Fe2.5+ on octahedral B sites resulting from the average

configuration due to a fast electron exchange between Fe2+

and Fe3+ on B sites. With increasing non-stoichiometry,

unpaired Fe3+ contributes to the apparent A-site area. The

ratio of the area of the two sextets of magnetite is the main

parameter allowing the evaluation of the average stoichiometry

of magnetite. This area ratio Fe2.5+/Fe3+ (R) is much higher

for the first two samples, FH1 and FH2 (Table 2). From this

value and according to the procedure described in a previous

study28 the average stoichiometry of the samples can be

calculated. Higher R value means higher stoichiometry as

shown in Table 2. Thus, these two samples have formulae

closed to the stoichiometric magnetite phase. Goethite

contribution can also be seen, which may represent partially

complete phase conversion into magnetite within the frame-

work of the mechanism proposed. In the other two spectra for

FH3 and FH4, the contribution of Fe3+ is significantly higher,

reflecting a more oxidized spinel oxide. This observation is

confirmed by the x value and the higher values of the isomer

shift d and hyperfine field Bhf, both explained by the increasing

amount of Fe3+ in the octahedral sites of the spinel structure

(Table 2). Samples FH5 and FH6 have broad spectra with a

slightly asymmetric shape characteristic of different isomer

shifts in agreement with a Fe3�xO4 spinel oxide. In so far as

the micro-state of the particle is concerned, three different

possibilities can be envisaged: a homogeneous particle composed

Fig. 6 Room temperature magnetization for FH1–FH4 cases.

Fig. 7 Mossbauer spectra for FH1–FH6 cases.

3250 | Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 This journal is �c the Owner Societies 2010

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online

of a partly oxidized magnetite, a core of stoichiometric

magnetite with a shell of maghemite,29 and a core of magnetite

surrounded by an oxidized layer with a gradient of oxidation

state. The stability of the hyperfine field of iron in hopping

state with such a variable R ratio allows us to decline the first

script (unpublished data). The third hypothesis is probably the

best one but it would be unrealistic to perform more improved

fitting procedure of Mossbauer spectra containing residual

mother compound. To get an idea of the contribution of the

oxidized fraction we can roughly evaluate the amount of iron

in a maghemite shell (second hypothesis). This amount is given

by the relation: [Femagh3+] = (Sp Fe3+ spectral area � 0.5*Sp

Fe2.5+ spectral area)/(Sp Fe3+ spectral area + Sp Fe2.5+

spectral area) and yields 27, 40, 81 and 81%, respectively for

FH1, FH2, FH3 and FH4.

Raman spectroscopy

Raman spectra of the FH1 to FH6 samples are shown in Fig. 8

in the low wave number region. These Raman spectra are

dominated by the Fe–O stretching modes of the Fe3�xO4

phase with the main and dominating broad 670 cm�1 band

of magnetite.

Magnetite has a spinel structure giving rise to five Raman

vibration modes with three main absorption bands, a prominent

one around 670 cm�1 and two bands of low intensity around

300 and 540 cm�1.30,31 Bands around 350, 500, and 700 cm�1

are reported to be the main vibrational modes of maghemite.30

Fe3�xO4 is an iron deficient form of magnetite with vacancies

mainly located in octahedral sites of the spinel structure. There

is no study in the literature concerning the impact of stoichio-

metry on Raman features. According to the available data, the

ratio between the main peak at 670 cm�1 and the broad

absorption band 300–600 cm�1 is a marker of the stoichio-

metry, along with the shoulder appearing around 710 cm�1.

Importantly, FH1 is the sample offering a stoichiometry not

far from magnetite. Goethite has two main peaks, 300 cm�1

and 390 cm�1, the first one being more efficient to detect

goethite in a multiphase sample.30,31 Goethite appears

obviously in the FH3 and FH4 samples.

Proposed mechanism

Literature reports22 show that in acidic media hexamine

undergoes slow thermal decomposition into formaldehyde

and ammonia. It was important therefore to analyze as

to whether slow decomposition of hexamine, presence of

ammonia or combination of ammonia and formaldehyde is

responsible for the formation of rods. Towards this end we

performed two experiments with ammonia only and ammonia

plus formaldehyde. These are discussed below.

Case 5: system FH5

In the XRD of Fig. 1 (FH5) corresponding to the case with

only ammonia as a hydrolyzing agent magnetite is found to be

Table 2 Fitted Hyperfine Interaction (Mossbauer) Parameters

Sample d/mm s�1 2e or D/mm s�1 Bhf/T Spectral area (%) R; 3 � x Assignment

FH1 0.30 � 0.01 — 49.2 � 0.1 35 � 2 0.93 � 0.10; 2.92 � 0.01 Sp Fe3+

0.66 � 0.05 — 45.9 � 0.4 33 � 2 Sp Fe2.5+

0.37 � 0.02 �0.27 � 0.04 30.3 � 0.3 21 � 1 Goethite0.34 � 0.01 0.64 � 0.02 — 11 � 1 Superpara

FH2 0.31 � 0.01 — 49.4 � 0.1 58 � 2 0.68 � 0.07; 2.87 � 0.01 Sp Fe3+

0.64 � 0.01 — 46.0 � 0.1 39 � 2 Sp Fe2.5+

0.36 � 0.04 0.56 � 0.06 — 3 � 1 Superpara

FH3 0.34 � 0.01 — 50.1 � 0.1 61 � 2 0.14 � 0.02; 2.73 � 0.01 Sp Fe3+

0.62 � 0.03 — 45.8 � 0.2 9 � 1 Sp Fe2.5+

0.38a � 0.01 0.24a � 0.02 40 to 4 Bal. goethite0.36 � 0.02 0.56 � 0.04 — 3 � 1 superpara

FH4 0.34 � 0.01 — 50.4 � 0.1 57 � 2 0.14 � 0.02; 2.73 � 0.01 Sp Fe3+

0.62 � 0.03 — 46.0 � 0.2 8 � 1 Sp Fe2.5+

0.39a � 0.01 0.26a � 0.02 40 to 2 Bal. Goethite

FH5 0.33a � 0.01 — 50 to 8–36.8b 95 — Sp0.33 � 0.01 0.60 � 0.02 5 Superpara

FH6 0.34a � 0.01 50 to 8–42.1b — — Sp

a The experimental data were fitted with an hyperfine field distribution with isomer shift and half width constrained to be equal. b Average,

Sp: Spinel.

Fig. 8 Raman spectra for the FH1–FH6 cases.

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 | 3251

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online

the main phase formed along with small contribution of

hematite. The TEM image for this case (Fig. 2 (FH5)) shows

faceted nanoparticles, but without any nanorods.

Case 6: system FH6

The XRD pattern (Fig. 1 (FH6)) for the sample synthesized

with ammonia and formaldehyde, peaks corresponding to

pure phase of magnetite are observed while the TEM of the

sample (Fig. 2 (FH6)) shows tiny uniformly sized nano-

particles of sizes between 10–15 nm. Importantly, as in the

FH5 case no nanorods are seen in this case as well.

We also performed synthesis using formaldehyde but no

compound formation took place, as expected. These experi-

ments cumulatively show that the decomposition products of

hexamine do not lead to a nanorod (or anisotropic structure)

formation. This strongly implies that the slow decomposition

of hexamine at elevated temperature and the consequent slow

rise in pH, are the important factors which contribute to the

anisotropic growth of the nanorods.

The possible mechanism involved in the formation of

nanorods can thus be inferred as follows: When the solution

containing ferrous ammonium sulfate (or FeSO4), ferric

chloride and hexamine is heated hexamine decomposes into

HCHO and NH3 the following reactions occur:

(CH2)6N4+6H2O - 4NH3 + 6HCHO (1)

NH3 + H2O - NH4+ + OH� (2)

Fe3+ + 3OH� - Fe(OH)3 (3)

Fe(OH)3 - FeOOH + H2O (4)

Fe2+ + 2OH� - Fe(OH)2 (5)

2FeOOH + Fe(OH)2 - Fe3O4 + 2H2O (6)

NH3 reacts with water, producing hydroxyl ions, which cause

a uniform rise in pH of the solution till the solubility limit. The

main advantage in this process is that uniform rise in the pH

prevents the occurrence of high local supersaturation, allowing

nucleation to occur homogeneously throughout the solution.

Since, the solubility product of Fe(OH)3 is much smaller than

that of Fe(OH)2, with the increase in the pH value, Fe(OH)3 is

first precipitated.32 Then, Fe(OH)3 gets converted into

FeOOH known as goethite, which has a circular form as its

principal crystal habit and it intrinsically leads to thin rod like

morphology.33 It is evident from the FTIR spectra of Fig. 3 as

well that we do get goethite as an intermediate. Separately, the

hydroxyl ions slowly produced in the solution react with Fe2+

ions to form Fe(OH)2. Finally, Fe(OH)2 would grow on

goethite and get converted into magnetite.

The role of hexamine as a shape directing agent is further

confirmed when a combination of NaOH and hexamine (FH3)

is used. In this system we got spherical nanoparticles showing

that the hexamine decomposes only in acidic condition and

presence of NaOH makes the system highly basic which

prevents the decomposition of hexamine.

Conclusions

In this work we have elucidated the mechanism of the magnetite

nanorod formation by hydrothermal method. It is found that

changes in the choice of iron precursors and combinations of

hydrolyzing/capping agents change the constitution, shape,

size and properties of phase(s) formed. Only when ferric

chloride, ferrous ammonium sulfate (FAS) and hexamine are

used, well defined nanorods are formed, though a small degree

of non-stoichiometry, related to oxidation phenomena, is also

noted. Interestingly, the decomposition products of hexamine

added as precursors do not lead to nanorod formation. We

conclude therefore that slow decomposition of hexamine at

elevated temperature and the consequent slow rise in pH are

the key factors responsible for the anisotropic growth of the

iron oxide system.

Acknowledgements

S. B. O. and B. H. thank CEFIPRA (Indo French project no.

3808-2) for funding support. S. B.O. also acknowledges

support from DST, Govt. of India for the award of the

Ramanujan fellowship.

References

1 D. J. Hornbaker, S.-J. Kahng, S. Misra, B. W. Smith,A. T. Johnson, E. J. Mele, D. E. Luzzi and A. Yazdani, Science,2002, 295, 828–831.

2 C. J. Murphy and N. R. Jana, Adv. Mater., 2002, 14, 80–82.3 N. Cordente, M. Respaud, F. Senocq, M. J. Casanove, C. Amiensand B. Chaudret, Nano Lett., 2001, 1, 565–568.

4 H. Nagano, Y. Machida, M. Iwata, T. Imada, Y. Noguchi,A. Matsumoto and T. Nagai, Int. J. Pharm., 1997, 147, 119–125.

5 A. K. Gupta and M. Gupta, Biomaterials, 2005, 26, 3995–4021.6 C. S. Lee, H. Lee and R. M.Westervelt, Appl. Phys. Lett., 2001, 79,3308–3310.

7 A. K. Gupta and A. Curtis, Biomaterials, 2004, 25, 3029–3040.8 D. K. Kim, Y. Zhang, W. Voit and K. V. Rao, J. Magn. Magn.Mater., 2001, 225, 30–36.

9 R. Takagi, J. Phys. Soc. Jpn., 1957, 12, 1212–1218.10 X. G. Wen, S. H. Wang, Y. Ding, Z. L. Wang and S. H. Yang,

J. Phys. Chem. B, 2005, 109, 215–220.11 Y. Y. Fu, R. M. Wang, J. Xu, J. Chen, Y. Yan, A. V. Narlikar and

H. Zhang, Chem. Phys. Lett., 2003, 379, 373–379.12 S. Y. Lian, E. B. Wang, Z. H. Kang, Y. P. Bai, L. Gao, M. Jiang,

C. W. Hu and L. Xu, Solid State Commun., 2004, 129, 485–490.13 S. Y. Lian, E. B. Wang, L. Gao, Z. H. Kang, D. Wu, Y. Lan and

L. Xu, Solid State Commun., 2004, 132, 375–378.14 Y. Li and H. Liao, Mater. Res. Bull., 1998, 33, 841–844.15 S. Takami, T. Sato, T. Mousavand, S. Ohara, M. Umetsu and

T. Adschiri, Mater. Lett., 2007, 61, 4769–4772.16 J. X. Wan, X. Y. Chen, Z. H. Wang, X. G. Yang and Y. T. Qian,

J. Cryst. Growth, 2005, 276, 571–576.17 K. He, C.-Y. Xu, L. Zhen and W.-Z. Shao, Mater. Lett., 2007, 61,

3159–3162.18 F. Zhang, Q. Jin and S.W. Chan, J. Appl. Phys., 2004, 95, 4319–4326.19 P. A. Haas, W. W. Pitt, Jr., S. M. Robinson and A. D. Ryon, Ind.

Eng. Chem. Prod. Res. Dev., 1983, 22, 461–466.20 A. Sugunan, H. C. Warad, M. Boman and J. Dutta, J. Sol-Gel Sci.

Technol., 2006, 39, 49–56.21 S. V. Bhagwat, S. Jouen, D. C. Kundaliya, H. Singh, T. Jagadale,

A. A. Athawale, S. Lofland, B. Hannoyer and S. B. Ogale,J. Nanosci. Nanotechnol., 2009, 9, 5823–5828.

22 H. Tada, J. Am. Chem. Soc., 1960, 82, 255–263.23 P. Cambier, Clay Miner., 1986, 21, 191–200.24 G. W. Poling, J. Electrochem. Soc., 1969, 116, 958–963.25 J. T. Keiser, C. W. Brown and R. H. Heidersbach, J. Electrochem.

Soc., 1982, 129, 2686–2689.

3252 | Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 This journal is �c the Owner Societies 2010

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online

26 D. Reinen, Ber. Bunsen-Ges. Phys. Chem., 1965, 69, 82.27 D. M. Sherman, Phys. Chem. Miner., 1985, 12, 161–175.28 S. Kale, S. M. Bhagat, S. E. Lofland, T. Scabarozi, S. B. Ogale,

A. Orozco, S. R. Shinde, T. Venkatesan, B. Hannoyer, B. Merceyand W. Prellier, Phys. Rev. B: Condens. Matter Mater. Phys., 2001,64, 205413–2054139.

29 T. J. Daou, J. M. Greneche, G. Pourroy, S. Buathong, A. Derory,C. Ulhaq-Bouillet, B. Donnio, D. Guillon and S. Begin-Colin,Chem. Mater., 2008, 20, 5869–5875.

30 D. L. A. de Faria, S. Venancio Silva and M. T. de Oliveira,J. Raman Spectrosc., 1997, 28, 873–878.

31 D. Neff, S. Reguer, L. Bellot-Gurlet, P. Dillmann andR. Bertholon, J. Raman Spectrosc., 2004, 35, 739–745.

32 S. Lian, E. Wang, Z. Kang, Y. Bai, L. Gao, M. Jiang, C. Hu andL. Xu, Solid State Commun., 2004, 129(8), 485–490.

33 R. M. Cornell and U. Schwertmann, in The Iron Oxides, WILEY-VCH Verlag GmbH and Co. KGaA, Weinheim, 2nd edn, 2003, ch.4, p. 64.

This journal is �c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 3246–3253 | 3253

Publ

ishe

d on

19

Febr

uary

201

0. D

ownl

oade

d by

Nat

iona

l Che

mic

al L

abor

ator

y, P

une

on 0

1/02

/201

4 05

:18:

03.

View Article Online