587
1 INTRODUCTION
Thereisaverylargenumberofwaystoprotectthe
materials from corrosion. If you analyze all the
compounds that are used in various countries for
protection of ferrous alloys corrosion by volatile
compounds,wecanseethattheyallcontainsodium
nitrite or nitrite ions belonging to the other
substances.
Protective properties of sodium nitrite NaNO
2,
found for a long time allowing to use it as an
inhibitoroftheatmosphericcorrosionof metals, for
theirpreservationandstorage,flushingconduits,etc.
Theprotectiveeffectoftheinhibitorsisduetotheir
moleculesorionsadsorbedonthemetalsurfaceand
thecorrosionrateisreducedcat
alytically.
To protect ferrous metals against atmospheric
corrosioninthe Russian Federation, for example, is
widelyusedinhibitedwrappingpaperthatcontains
an inhibitor UNI (a mixture of sodium nitrite and
hexamine),(GOST16295‐931993).Sodiumnitrite‐a
non‐volatile inhibitor, urotropine‐volatile and low
toxicity inhibitor. Due to direct contact (sodium
nitrite) and release fumes (urot
ropine) anticorrosive
paperprotectsthepackagedgoodsfromcorrosion.
One can use viscous solutions of sodium nitrite
for long term storage. This is a contact inhibitor
depositedonthesurfaceoftheobject.Anaddingof
sodium nitrite into the aqueous solutions, that is a
substance increasing the viscosity significant
ly
increasestheefficiency andlengthensthetimeforthe
shields when storing products in any climatic
conditions.
This prevents drying of a solution of sodium
nitriteandfallingofsaltcrystalsfromthesurfaceof
themetalandreduces the refluxing solution due to
liquefact
ionathighhumidity.
However, sodium nitrite, being a good inhibitor
for steel, destroys the solder, i.e., in systems
containing a number of heterogeneous
electrochemical attitude of metals and alloys (steel,
zinc, brass, solder, cast iron, aluminium), cannot be
appliedsingly.
Investigation of Corrosion Inhibitors by Nuclear
Quadrupole Resonance Relaxometry Method
N.Ya.Sinyavsky
BalticFishingFleetStateAcademy,ImmanuelKantBalticFederalUniversity,Kaliningrad,Russia
I.G.Mershiev
ImmanuelKantBalticFederalUniversity,Kaliningrad,Russia
I.P.Korneva
BalticFishingFleetStateAcademy,ImmanuelKantBalticFederalUniversity,Kaliningrad,Russia
ABSTRACT: The changes taking place with the corrosion‐resistant coating, but not the state of the surface
subjectedtocorrosionareinvestigatedinthispaperincontrasttotraditionalapproaches.Weusedthemethod
of nitrogen relaxometry NQR and multi‐exponential inversion of decay of longit
udinal and transverse
components of the nuclear magnetization is applied for the first time for this purpose. The results of
experimental studies of changes in the distributions of spin‐spin and spin‐lattice relaxation of crystallite
powderofsodiumnitriteandurotropin,themixtureofwhichisusedasacorrosioninhibitorofferrousmeta
ls
areconsidered.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 9
Number 4
December 2015
DOI:10.12716/1001.09.04.17
588
In the paper (Hassan et al. 2013), sodium nitrite
was studied as an inhibitor for the protection of
reinforcing steel in concrete from corrosion. The
purposeofthisstudywasanattempttounderstand
the mechanism by which sodium nitrite contributes
to reducing the rate of corrosion on the surface of
carbon steel, depending on pH and the presence of
chlorideinthesystem.Inthelit
erature(Rosenberg&
Gaidis1979)ithavesuggestedthatnitriteundergoes
achemicalreactionwiththedivalentiron:
OHOFeNONOOHFe
2322
2
2222
(1)
ThisreactionformsthebarrierfilmofoxideFe
2O3
on the metal surface. The film protects from
corrosion.Thecorrosionratelargelydependsonthe
pH of the solution. The presence of chlorides
increasesthereactionrateofcorrosionbyactingasa
catalyst. Nitrite ions act as inhibitors by increasing
the rate of the barrier iron oxide film formation.
Increasing temperature leads to the desorption of
adsorb
ed sodium nitrite on the surface of the steel
andreducestheeffectivenessoftheinhibitor.
Theaimofthisworkwastostudychangesinthe
relaxation times distribution of nuclear quadrupole
resonance (NQR) in microscopic inhibitor after its
contactwiththesurfaceofinhibitedbla
ckmetal. This
approach is original since the change of the metal
surfacestatebycorrosioninthepresenceofinhibitor
isusuallyinvestigated.
2 EXPERIMENTALSTUDY
We have used nuclear quadrupole resonance as a
researchmethodbeca use relaxationtimes,spectrum
and shape of the NQR lines are very sensit
ive to
small changes in the environment of the nuclei.
Unlike nuclear magnetic resonance (NMR), nuclear
quadrupoleresonancefrequenciesaredeterminedby
the electric interactions, so the method allows not
indirectly, but directly to study the electric field at
the nuclei of atoms. NQR spectroscopy data are of
fundamental importance for understanding the
changesintheelectronicdistribution,theobservation
of physica
l phenomena occurring in crystals at the
microlevel.
Relaxationtimes express molecular dynamics, as
wellasprovideinformationaboutthemicrostructure
of matter. For homogeneous samples the signals of
nuclear quadrupole resonance recedes by the same
exponent.Inthi
scase,onlytherelaxationtimecanbe
preciselydeterminedbya linearapproximation.The
relaxation time distribution is continuous for more
complexsamples,molecularcrystalswithimpurities
micropowders, porous media due to various
intermolecular and many exponential inversion is
required for definition of the relaxation times
distribution.
LongitudinalNQRrelaxationti
mesaredefinedby
spin‐lattice interaction with the transfer of energy
fromthenuclearspinstomobilelattice,relaxationof
thetransversenuclearmagnetizationdependsonthe
spin‐spin interactions and due to the
inhomogeneities of crystallite lattice. Magnetization
of the nuclei in the different materials differing in
density, molecular structure and mobilit
y in
dependence on the environment is restored to the
individual time constant. NQR signals relaxation in
solidsareverydependentonthedegreeofordering
in the sample. Amorphous samples generally relax
fasterthancrystallinesamples.
Itwasfoundpreviouslythatthelongitudinaland
transverse relaxation ti
mes of nuclei in the fine
particles are reduced in comparison with the
corresponding samples in large samples. It is
assumed that this is caused by magnetization
diffusionfromthebulktothecrystallitesurface.Itis
known that the phonon spectrum of atoms at the
crystalsurfacediffersfromatomsinthebulk,dueto
thehighconcentrationofl
atticedefects,dislocations,
amorphouscharacter,causedbyplasticdeformation
during the milling process, or the influence of the
environmentsurroundingthecrystal.Theresultisa
strongconnectionofthesurfacespinswiththelattice
duetofluctuationscausedby theincreased number
ofdegreesoffreedom.
Experimental measurements were performed on
the NQR spectrometer Tecmag Apollo with TNMR
software (Fig. 1). The pulse sequence recovery
inversion 180° –τ
– 90° was used to measure the
spin‐latticerelaxationtime.Carr‐Purcell‐Meiboom
‐ Gill (CPMG) sequence was used to measure the
spin‐spinrelaxationti
meT
2.T2
*
relaxation timewas
measured from the shape of the Hahn echo signal.
Finally, the sequences with spin‐locking pulse of
variable duration was used to measure the spin‐
lattice relaxation in the rotating coordinate system
T
1
.Allpulsesequencesarewellknownandrequire
nospecialdescription.
The program RILT (Regularized Inverse Laplace
Transform)wasusedfortheinversionoftheLaplace
transformdescribedintheIari‐GabrielMarinowork
(Marino 2004). The desired array of relaxation time
distribution f(T) is the inverse Laplace transform of
the ti
me signal exponential decays measured by an
arrayofS(t).Arrayiscalculatedbytheregularization
using the least squares method. In the process of
calculation was used from 50 to 200 iterations.
Inversion of the Laplace transform have been used
effectively by us previously (Sinyavsky et al. 2014,
Dolinenkov et al. 2014) in the study of micro‐
composite and porous ma
terials and phase
transitions.
Figure1.Tecmag‐ApolloNQRspectrometerandinterface
ofTNMRsoftware.
589
Themethodofsamplepreparationwasfollowing.
Small shavings of iron was placed in a saturated
aqueoussolutionofinhibitorUNI(50%NaNO
2and
50%C
6H12N4)fora1.5month.Afterthat,thesolution
wasevaporated,crystallized powder wasdried and
ground in a mortar. Iron particles were removed
withamagnet.NQRmeasurementsofthesamplein
thepresenceofmetalparticlesareimpossibledueto
thefallofthequalityfactoroftheworking
circuitof
spectrometer sensor.All measurements were
performedatroomtemperatureT=297K.
3 RESULTSANDDISCUSSION
Studieshaveallowedtoestablishthatthe14NNQR
linewidthoftheurotropineinthesampleincontact
with the iron powder, practically unchanged
comparedwiththecontrolsample(770
±20Hz).The
sameistrueforsodiumnitrite,widthforthelineν‐=
3.603MHzisintherangeof280±30Hzfortestand
controlsamples.
The relaxation times distribution of the spin‐
latticeT
1inurotropineandsodiumnitriteintheUNI
inhibitor after contact with iron and in the control
samplesareshowninFig.2andFig.3,respectively.
Distributionsare of a multimodal nature even for a
control sample, which was not in contact with the
iron.ThetimesdistributionT
1inthesamplesofpure
urotropine and pure sodium nitrite are single‐
bimodal,inaccordancewiththetheoreticalmodelof
the magnetization diffusion shown in the article
(Sinyavsky 2014). It is shown in the paper that the
modaldistributionofrelaxationtimesisdetermined
bythevalueofthediffusion
coefficientofthenuclear
magnetizationdistributionandlocalinhomogeneities
nearthesurfaceofthecrystallite.Apparently,mutual
influenceofNaNO
2andC6H12N4onthesurfaceofthe
crystallitesofbothsubstancesleadstomultimodality.
However, as can be seen from Fig. 2 ‐3 time
distribution T
1 in urotropine and for the relaxation
timesdistributionT
1inthesodiumnitritesignificant
changesareobservedtheninhibitorisinthecontact
with iron. This is most likely due to change of the
atomsmobilityonthesurfaceofthecrystallites.Ifthe
mobilityoftheatomsonthesurfaceofthesolidbody
is higher, then likely
spin relaxation time on the
surface is shorter. It seems, that processes of spin‐
spin diffusion have an important role in the
relaxation times distribution, leading to the transfer
ofmagnetizationfromthemassofcrystalliteonthe
surface.
In the literature, so far of the research results of
changes occurring
on the surface of the crystal
inhibitorbycontactwiththesurfaceofferrousmetal,
thereisno.
Figure2. Distribution of spin‐lattice relaxation times
14
N
NQRinNMTofUNIinhibitor:a‐controlsample,b‐isthe
sampleaftercontactwithiron.
Thereare three peaksinthe spin‐spinrelaxation
times distribution in investigated sample and in
controlsampleofUNIinhibitor(Fig.4),incontrastto
pure sodium nitrite the time distribution T
2 is
bimodal (Sinyavsky et al. 2014). The peak in the
region of 100 ms is not shifted in comparison with
thecontrolsampleandiscaused,apparently,bythe
effectofurotropinetothesurfaceofNaNO
2crystals
thatcrystallizefromaqueoussolutioninthepresence
ofC
6H12N4.TheinfluenceofironoxideFe2O3onthe
surfaceofthesodiumnitritecrystalsleadstotheshift
ofbothshortpeaks.
Figure3.TherelaxationtimesdistributionT1inNaNO2of
UNIinhibitor:a‐thecontrolsample,b‐thesampleafter
contactwithiron.
590
Figure4.TherelaxationtimesdistributionT2inNaNO2of
UNIinhibitor:a‐thecontrolsample,b‐thesampleafter
contactwithiron.
Therelaxationtimesdistributionofspin‐latticein
theUNIinhibitoronnucleiofnitrogensodiumnitrite
atfrequencyν
‐= 3.603MHzat thetemperatureT =
297Kaftercontactwiththeironforabout1.5months
andinthecontrolsampleareshowninFig.5.
Spin‐latticerelaxationintherotatingframe(T
1ρ)is
the mechanism through which the excited
magnetizationvectorisdecaysundertheinfluenceof
thespin‐lockingRFirradiation.T
1ρ‐filtrationisused
toseparatesignalsfromamixtureofamorphousand
crystallinematerialinNMRmethods.
Asseenfromthefigure,thetimedistributionsT
1
are unimodal, and T
1
value in the inhibitor
contacting with the iron decreases 3‐4 times
comparedtothecontrolsample.
Figure5.TherelaxationtimesdistributionT1
inNaNO2of
UNIinhibitor:a‐thecontrolsample,b‐thesampleafter
contactwithiron.
4 CONCLUSION
Thus, unlike traditional approaches, when studied
surfacestatecorrodeinthepresenceofinhibitor,we
examine the changes occurring inhibitor with itself
after its contact with the protected surface. For the
first time for this purpose we used the NQR
relaxometry method and the Laplace transform
inversion method
for the relaxation times
distributionofthenuclearmagnetization.
The results of experimental studies of spin‐spin
and spin‐lattice relaxation time distribution in
crystallitepowderofsodium nitriteand urotropine,
which mixture is used as a corrosion inhibitor of
ferrousmetals.Itwasfoundthatthesedistributions
areverysensitive
tothestateofthelayersurfaceof
theinvestigatedinhibitorcrystallitesandmaterialof
theirenvironment.Itisshown(theUNIinhibitorwas
the example) that the proposed method can be
successfully used to study the inhibitory properties
of drugs used to prevent corrosion in different
conditions.
Theresults
obtainedinthispapercanbeapplied
tothestudyofcorrosioninhibitors,inordertofind
themosteffectivemeansofprotectingofthecooling
systems and of water ballast tanks of sea water on
ships and in the floating docks, of various metal
constructions in the water, in ports,
during storage
andtransportationofthemetalproducts,etc.
REFERENCES
Dolinenkov P., Korneva I. & Sinyavsky N. 2015. The
distributionchangeofrelaxationtimesin
35
ClNQRfor
phase transitionsin p‐Dichlorobenzene. Applied
MagneticResonance46(1):(inprint).
GOST 16295‐93. 1993. Interstate standard. Anticorrosive
paper (in Russian). http://www.germesagro.ru/gosti/
gost1629593.php
MarinoI.‐G.2004.RegularizedInverseLaplaceTransform,
http://www.mathworks.com/matlabcentral/fileexchang
e/6523‐rilt/content/rilt.m
HassanA.R.,IsmaelA.K.,HassanF.N.2013.Sodiumnitrite
as inhibitors for protection of
rebar against corrosion.
Al‐MuthannaJournalforEngineeringSciences2(1):82‐96.
RosenbergA.H.andGaidisJ.M.1979.Themechanismof
nitriteInhibitionofchlorideattackonreinforcingsteel
inalkalineaqueousenvironments.MaterialsPerformance
18(11):45–48.
SinyavskyNikolay,MozzhukhinGeorgyV.&Dolinenkov
Philip. 2014. Size Effect
in 14N Nuclear Quadrupole
Resonance Spectroscopy, In T. Apih et al. (eds.),
Magnetic Resonance Detection of Explosives and Illicit
Materials, NATO Science for Peace and Security Series B:
PhysicsandBiophysics:69‐76.Dordrecht:Springer.
SinyavskyN.,DolinenkovP.,KupriyanovaG.2014.TheT
1
and T
2 Relaxation Times Distribution for the
35
Cl and
14
NNQRinMicro‐compositesandinPorousMaterials.
AppliedMagneticResonance45:471–482.
SinyavskyN.Ya.2014.EffectoftheNuclearMagnetization
Diffusion on the Relaxation Time Distribution in
Microcrystals.RussianPhysicsJournal57(8):1149‐1151.