Thermal Analysis

Application
Booklet

Thermal Analysis of Petrochemicals

Application Examples
Application Examples
Thermal Analysis

Thermal Analysis of Petrochemicals

This application handbook presents selected application examples. The experiments were conducted with the
utmost care using the instruments specified in the description of each application. The results were evaluated
according to the current state of our knowledge.
This does not however absolve you from personally testing the suitability of the examples for your own methods,
instruments and purposes. Since the transfer and use of an application is beyond our control, we cannot of
course accept any responsibility.
When chemicals, solvents and gases are used, general safety rules and the instructions given by the manufac-
turer or supplier must be observed.

® TM All names of commercial products can be registered trademarks, even if they are not denoted as such.

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 3

Preface

Thermal analysis is one of the oldest analysis techniques. Throughout history, people have used simple heat
tests to determine whether materials were genuine or fake.
The year 1887 is looked upon as the dawn of present-day thermal analysis. It was then that Henry Le Chatelier,
the famous French scientist, carried out his first thermometric measurements on clays. Just a few years later in
1899, the British scientist William Roberts-Austen performed the first differential temperature measurements
and so initiated the development of DTA. Commercial instruments however, did not appear until the early
1960s. Since then, thermal analysis has undergone fifty years of intense development.
Thermal analysis is the ideal technique for determining material properties and transitions and for character-
izing materials.

This application booklet contains a number of different thermal analysis application examples taken from the
petrochemical industry.
The applications were previously published in various issues of UserCom, our bi-annual application magazine
(www.mt.com/ta-usercoms), or in one of our application handbooks (www.mt.com/ta-handbooks).

From crude oil to gasoline, diesel, liquid petroleum gas, lubricants, polymers and many other end products,
petrochemicals are ever present in our lives. The number of standards and regulations for analysis further
reflects the importance and variety of petrochemical raw materials and products. In today’s competitive en-
vironment, expectations are high when it comes to meeting quality and regulatory requirements, improving
yield, and reducing downtime, corrosion and waste.
Thermal Analysis is therefore an extremely useful tool for the quality control and analysis of petrochemical
compounds.

We hope that the applications described here will stimulate your interest and make you aware of the great
potential that thermal analysis methods have in the field of petrochemistry.

Dr. Angela Hammer and the editorial team of the METTLER TOLEDO materials characterization group:

Nicolas Fedelich
Samuele Giani
Dr. Elke Hempel
Ni Jing
Dr. Melanie Nijman
Dr. Rudolf Riesen
Dr. Jürgen Schawe
Dr. Markus Schubnell

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 5

 1. Introduction 8
1.1 About this Handbook 8
1.2 – 1.11 Important Thermal Analysis Techniques 8
1.12 Application Overview 9
2. Characterization of petroleum products with DSC 10
2.1 Introduction 10
2.2 Characterization of petroleum products with DSC 10
2.3 Evaluation of DSC curves of petroleum products 11
2.4 Conclusions 12
2.5 References 12
3. Oxidative stability of petroleum oil fractions 13
3.1 Introduction 13
3.2 Samples 13
3.4 Conclusions 13
3.5 References 13
4. Determination of the Noack evaporation loss of lubricants by TGA 14
4.1 Abstract 14
4.2 Introduction 14
4.3 The Noack evaporation loss test according to ASTM D6375 14
4.4 Performing a Noack test 15
4.5 Conclusions 15
4.6 References 15
5. R
 apid thermogravimetric analysis of coal 16
5.1 Introduction 16
5.2 Speeding up the TGA procedure 16
5.3 Experimental details 16
5.4 Results 17
5.5 Conclusions 17
5.6 References 18
6. Determination of oxidation stability by pressure-dependent OIT measurements 19
6.1 Introduction 19
6.2 Experimental details 19
6.3 Pressure dependence of OIT 19
6.4 Interpretation of the results 20
6.5 Assessment of experimental results 21
6.6 Conclusions 22
7. Soot analysis study of engine oil samples 23
7.1 Purpose of analysis 23
7.2 Interpretation 23
7.3 Evaluations 23
7.4 Conclusions 23
8. Melting behavior of oil/wax mixtures 24
8.1 Purpose of analysis 24
8.2 Evaluation 24
8.3 Conclusion 24
9. S
 tandards for petrochemicals with respect to thermal analysis 25

10. For More Information 26

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 7

 1. Introduction
1.1 About this Handbook 1.3 DTA DSC allows you to detect endothermic
This handbook shows how thermal anal- Differential Thermal Analysis and exothermic effects, measure peak ar-
Introduction

ysis techniques can be used to analyze In DTA, the temperature difference be- eas (transition and reaction enthalpies),
polymers and in particular to study the tween the sample and an inert reference determine temperatures that character-
behavior of thermoplastics, thermosets substance is measured as a function of ize a peak or other effects, and measure
and elastomers. temperature. The DTA signal is °C or K. specific heat capacity.
Previously, the thermocouple voltage in
The chapters describe many interesting millivolts was displayed. 1.6 TGA
examples that illustrate the power of Thermogravimetric Analysis
thermal analysis for measuring physical 1.4 SDTA TGA measures the weight and hence the
properties, different types of transitions, Single DTA mass of a sample as a function of tem-
aging, the effect of fillers and additives, This technique was patented by METTLER perature. Previously, the acronym TG was
and the influence of production condi- TOLEDO and is a variation of classical used for this technique. Nowadays, TGA
tions. DTA that is particularly advantageous is preferred in order to avoid confusion
when used in combination with thermo- with Tg, the glass transition temperature.
The experiments were performed using gravimetric analysis. The measurement TGA allows you to detect changes in the
three different types of plastic materials, signal represents the temperature differ- mass of a sample (gain or loss), evaluate
namely a thermoplastic (PET), a thermo- ence between the sample and a previously stepwise changes in mass (usually as a
set (KU600), and an elastomer (W001). measured and stored blank sample. percentage of the initial sample mass),
and determine temperatures that char-
1.2 Important Thermal Analysis DTA and SDTA allow you to detect endo- acterize a step in the mass loss or mass
Techniques thermic and exothermic effects, and to gain curve.
The following sections give a brief expla- determine temperatures that character-
nation of some of the important thermal ize thermal effects. 1.7 EGA
analysis techniques. The four main tech- Evolved Gas Analysis
niques, DSC, TGA, TMA and DMA used in 1.5 DSC EGA is the name for a family of tech-
this handbook are often complementary. Differential Scanning Calorimetry. niques by means of which the nature
Sometimes however, only a combination In DSC, the heat flow to and from a sam- and/or amount of gaseous volatile prod-
of all four techniques provides a full in- ple and a reference material is measured ucts evolved from a sample is measured
sight into the sample. as a function of temperature as the sam- as a function of temperature. The most
ple is heated, cooled or held at constant important analysis techniques are mass
This is illustrated in Figure 1 which temperature. The measurement signal is spectrometry and infrared spectrometry.
shows the measurement of a sample of the energy absorbed by or released by the EGA is often used in combination with
polyamide 6 using DSC, TGA and TMA. sample in milliwatts. TGA instruments because TGA effects
involve the elimination of volatile com-
Figure 1.
The techniques pounds (mass loss).
used to measure
polyamide 6 show
1.8 TMA
different thermal
effects. DSC: melt- Thermomechanical Analysis
ing peak of the TMA measures the deformation and di-
crystalline part;
TGA: drying and
mensional changes of a sample as a
decomposition step; function of temperature. In TMA, the
TMA: softening un- sample is subjected to a constant force,
der load.
an increasing force, or a modulated
force, whereas in dilatometry dimen-
sional changes are measured using the
smallest possible load.

Depending on the measurement mode,

TMA allows you to detect thermal effects
(swelling or shrinkage, softening, change
in the expansion coefficient), determine

8 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

temperatures that characterize a thermal The most important results are tempera- cal transmission by means of hot-stage
effect, measure deformation step heights, tures that characterize a thermal effect, microscopy or DSC microscopy. Typical
and to determine expansion coefficients. the loss angle (the phase shift), the me- applications are the investigation of crys-
chanical loss factor (the tangent of the tallization and melting processes and
1.9 DMA phase shift), the elastic modulus or its polymorphic transitions.
Dynamic Mechanical Analysis components the storage and loss moduli,
In DMA, the sample is subjected to a si- and the shear modulus or its components 1.11 TCL
nusoidal mechanical stress. The force the storage and loss moduli. Thermochemiluminescence
amplitude, displacement (deforma- TCL is a technique that allows you to ob-
tion) amplitude, and phase shift are 1.10 TOA serve and measure the weak light emis-
determined as a funtion of temperature Thermo-optical Analysis sion that accompanies certain chemical
or frequency. DMA allows you to detect By TOA we mean the visual observation reactions.
thermal effects based on changes in the of a sample using transmitted or reflect-
modulus or damping behavior. ed light, or the measurement of its opti-

1.12 Application Overview

Table 1.
Property or application DSC DTA TGA TMA DMA TOA TCL EGA
Application over-
Specific heat capacity ••• • view showing the
thermal analysis
Enthalpy changes, enthalpy of conversion ••• • techniques that can
Enthalpy of melting, crystallinity ••• • be used to study
particular properties
Melting point, melting behavior (liquid fraction) ••• • • ••• or perform certain
applications.
Purity of crystalline non-polymeric substances ••• ••• •
Crystallization behavior, supercooling ••• • •••
Vaporization, sublimation, desorption ••• • ••• ••• •••
Solid–solid transitions, polymorphism ••• ••• • •••
Glass transition, amorphous softening ••• • ••• ••• •
Thermal decomposition, pyrolysis, depolymerization, • • ••• • • •••
and degradation
Temperature stability • • ••• • • •••
Chemical reactions, e.g. polymerization ••• • • •
Investigation of reaction kinetics and applied kinetics ••• • ••• •
(predictions)
Oxidative degradation, oxidation stability ••• ••• ••• • •••
Compositional analysis ••• ••• •••
Comparison of different lots and batches, competitive ••• • ••• • • ••• • •••
products
Linear expansion coefficient •••
Elastic modulus • •••
Shear modulus •••
Mechanical damping •••
Viscoelastic behavior • •••
••• means “very suitable”, • means “less suitable”

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 9

 2. Characterization of petroleum products with DSC
2.1 Introduction Distillation of the crude oil yields various 1. Crystallized material separates out
Conventional fuel is obtained exclusively fractions, which are classified as follows: forming a sediment. This is often a
Application

from petroleum or crude oil. Petroleum low boiling fractions, e.g. gasoline (pet- problem, especially for the storage of
is primarily a mixture of 6 different rol), aviation gasoline, naphtha; higher diesel and heating oils.
classes of substances. The composition boiling fractions, e.g. fuel or heating oil 2. Crystallized material is retained in
of the mixture is specific to the region and diesel; and high boiling fractions filters, which can lead to blockages.
where the oil occurs and consists of (heavy oil and lubricating oils). The resi- 3. Bitumen (asphalt) products are mainly
• straight-chain n-alkanes (CnH2n+2) with due after distillation is known as bitumen used for surfacing roads. Crystalli-
molar masses between 16 and 300 g/mol (or asphalt). zation causes the surface to become
• branched-chain alkanes (iso-alkanes) brittle and results in the formation of
• cycloalkanes In the liquid state, the distillate appears cracks.
• aromatics macroscopically as a single-phase mix-
• sulfur-containing compounds ture. On cooling, crystals are formed, i.e. Hydrocarbon distillates consist primarily
• polycyclic and heterocyclic resins as a multiphase mixture is obtained. The of complex hydrocarbon compounds and
well as bitumens with molar masses separation of crystalline material is unde- crystallizable fractions. The former are
of about 1000 g/mol. sirable and leads to a number of problems: partially liquid at room temperature and
exhibit a glass transition at low tempera-
Figure 1.
DSC curves of tures. The glass transition temperatures
different petroleum of the liquid constituents depend on the
distillates.
petroleum distillate. Typical values are
–30 °C for bitumen, –130 °C for diesel
and –150 °C for gasoline.

The proportion of the crystallizable frac-

tions is between 0% and 10% for bitumen,
between 5% and 25% for fuel oil and
up to 40% for crude oil. The chemical
structure of the crystals depends on the
distillate.

With fuel oils, n-alkanes with 10 to 28

carbon atoms crystallize out, with bitu-
men n-alkanes with 20 to 60 carbon at-
oms; and with crude oil n-alkanes with
5 to 60 carbon atoms. Lightly branched
Figure 2.
Evaluation of a DSC iso-alkanes and cycloalkanes are also
curve of diesel oil. present.

2.2 Characterization of
petroleum products with DSC
Petroleum products are usually char-
acterized by their glass transition tem-
perature and their melting behavior.
The measurement of these parameters
is easily performed by DSC. A typical
temperature program for the analysis of
petroleum derivatives begins by cooling
the sample at 10 K/min from room tem-
perature to 100 °C (for heavy hydrocar-
bon compounds) or to –150 °C (for light
hydrocarbon compounds, e.g. kerosene,
gasoline).

10 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

Afterward the sample is heated linearly In practice, a value of 200 J/g is has prov- For the evaluation of the corresponding
at 5 K/min up to final temperatures of en best. With crude oils and heavy oils, a DSC curve, one distinguishes between the
typically 50 °C (for fuel oils such as light value of 160 J/g is recommended below following characteristic temperatures:
heating oil or diesel), 80 °C (crude oil), 30 °C and a value of 200 J/g above 30 °C. • the turbidity (cloud) point, with crude
100 °C (heavy oil) and 120 °C (bitu- For a number of practical reasons, the and heavy oils often also called the wax
men). Figure 1 shows the corresponding problems mentioned at the beginning in appearance temperature (WAT), cor-
heating curves for different samples. Vari- connection with crystals separating out responds to the temperature at which
ous effects can be observed. on cooling are particularly important. crystallization begins (ASTM D2500).
• the CFPP, Cold Filter Plugging Point,
The marked increase of the heat capacity With crude oils and heavy oils, it is corresponds to the temperature below
at low temperatures (step in the heat best to measure the crystallization in which all crystallizable material has
flow curve) is caused by the glass transi- the temperature range 80 °C to –20 °C crystallized (EN 116).
tion. Very often, part of the sample then at a cooling rate of 2 K/min; and with • the flow point (FP) is the temperature
crystallizes out (usually iso-alkanes), medium heavy fuel oils, between 25 °C at which the viscosity of the sample
causing an exothermic peak. to –30 °C at a cooling rate of 0.5 K/min. is so high that it no longer flows
(ASTM D97).
The melting of the various crystals can In such experiments, a more or less pro-
lead to several relatively broad, endo- nounced exothermic peak is observed that To evaluate the crystallization peak,
thermic peaks. The shape of the peak shows the course of the crystallization. a horizontal or tangential baseline
mirrors the size and weight distribution
Figure 3.
of the crystals and is characteristic of a Typical
particular distillate or particular crude oil. crystallization
behavior
of diesel oil.
2.3 Evaluation of DSC curves
of petroleum products
The crystalline components are embed-
ded in the liquid matrix. The matrix
is characterized by the glass transition
temperature Tg and the step height of
the change of the specific heat. The
glass transition temperature correlates
well with the average mole mass of the
matrix.

The percentage amount of the crys-

tallized fractions can be calculated
by dividing the measured heat of
fusion by the (in principle temperature-
Figure 4.
dependent) melting enthalpy ∆H(T) of a Typical
fictive, fully crystallized sample. crystallization
behavior
of diesel oil.
For the compounds considered here,
a linear baseline can be used for the
determination of the peak area. This
begins at about Tg +30 K (Ti) and ends
at about 10 K after the end of the melting
process (Tf ). The melting enthalpy of the
crystallized material can be estimated in
the following way.

For medium distillates (gasoline, heat-

ing oil), ∆H(T) can be described by a
third order polynomial [1]. It is sufficient
to use a constant value of 160 J/g. With
bitumen, the melting enthalpy is larger.

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 11

 is drawn on the left side of the curve. 0.45% of the crystallizable material has 2.4 Conclusions
A value of 200 J/g is assumed for the crystallized out (Tc(0.45%)) and the CFPP DSC measurements allow a rapid and
crystallization enthalpy. determined according to EN 116: reliable characterization of extremely
different types of petroleum products.
Example 1: Tc(0.45%) = 1.01⋅TCFPP EN 106 – 0.85. The glass transition temperature and
Application

light petroleum distillate (Fig. 3) melting behavior provide important in-

The turbidity point is taken to be the For the determination of the flow point formation on the quality of petroleum
onset temperature of the crystallization we found the optimum correlation to be: derivatives.
peak (Tonset). The turbidity point defined
in this way can be reproducibly measured Tc(1%) = 1.02·Tflow point ASTM – 0.28 In addition, the origin of a sample of
with an accuracy of ±0.5 K. unknown petroleum can be determined
[2]. Here, Tc(1%) is the temperature at because the measured curves are char-
The values obtained with this method are which 1% of the crystallizable fractions acteristic of the petroleum, i.e. they are
slightly lower than those determined us- has crystallized out. “fingerprints”. Cooling experiments yield
ing the ASTM standard method (TASTM). important practical information on the
Example 2: crystallization behavior of petroleum
The measurement of 50 different light heavy and crude oils (Fig. 4) derivatives.
distillates led to the following correlation: The determination of the turbidity point
for heavy and crude oils is performed in a 2.5 References
WAT = Tonset = 0.98·TASTM –3.6 similar way to the light petroleum prod- [1] F. Bosselet, Thèse Saint Etienne
n°008 (1984)
ucts. If the flow point is below 0 °C, the [2] P. Claudy, J.M. Létoffé, B. Neff, B.
For the determination of the CFPP val- crystalline content after cooling is only Damin, FUEL, 65 (1986) 861-864
[3] J.M. Létoffé, P. Claudy, M.V. Kok,
ue, the analysis of 40 light petroleum about 2 mass%. The behavior of the sam-
M. Garcin, J.L. Volle, FUEL, 74 (6)
products gave the following correlation ple is therefore for the most part deter- (1995) 810-817
between the temperature at which mined by the noncrystalline matrix [3].

12 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

3. Oxidative stability of petroleum oil fractions
3.1 Introduction heat of the sample and no other thermal baseline before oxidation and the tangent
The determination of the oxidative in- effects are observed before the beginning at the steepest part of the DSC curve:
duction temperature is a rapid method of oxidation. The full curve for light diesel
for assessing the stability of petroleum oil oil (LGO) is shown in the upper diagram; 3.4 Conclusions
fractions. The same method can be used the change in the heating rate at 150 °C The stability of the various fractions is
to measure the effectiveness of stabilizers. is marked with a vertical line. The lower very different. The DSC curves (Figure 1)
diagram shows the measurements of all also show differences at the beginning
Besides this, aging processes can also the samples from 150 °C onwards. of oxidation. Similar fractions, however,
be investigated. Standardized isother- show almost identical DSC curves with
mal methods are often used instead of The curves are arranged so that they can practically the same onset temperatures
dynamic methods (e.g. ASTM D 5483 and be seen clearly. The tangents used for the (e.g. VGO1 and VGO2). This makes it very
ASTM E 1858). evaluation are drawn for the samples of easy to classify the fractions according
lowest and highest stability. The abbre- to their stability. Long-term stability at
Analysis under high pressures of oxygen viations refer to the petroleum fractions low temperatures can, however, only be
prevents the vaporization of volatile com- men tioned above. predicted with kinetic experiments.
ponents and increases the rate of oxida-
tion, thereby shortening the measure- Evaluation 3.5 References
ment time [1, 2]. The oxidative stability is compared by [1] A .T. Riga and G. H. Patterson,
Eds., Oxidative Behavior of Materials
determining the temperature at which by Thermoanalytical Techniques,
3.2 Samples oxidation begins (onset). This value is ASTM STP 1326, American Society
for Testing and Materials, 1997
Diesel oils of the following petroleum considered to be characteristic of the
[2] H. Kopsch, Thermal Methods in
fractions: Light (LGO), Light Cycle sample. It is measured as the point of in- Petroleum Analysis, VCH Verlags-
(LCGO), Light Vacuum (LVGO), Vacuum1 tersection of the horizontal tangent of the gesellschaft, Weinheim,1995.
(VGO1), Vacuum2 (VGO2), Vacuum3
Figure 1.
(VGO3) and Kerosine. Oxidative stability of
vacuum gas oil.
Information expected
Comparison of the products with respect
to their stability in oxygen.

3.3 Measurement parameters

• Measuring cell: DSC27HP / TC15
• Crucible: Aluminum 40 µl, with
pierced lid (1 mm hole)
• DSC measurement: Put the measuring
cell under an atmosphere of oxygen
and heat from 40 °C to 150 °C at
20 K/min (to save time), then
continue up to 350 °C at 5 K/min.
The measurement is automatically
terminated when the value of the
exothermic DSC signal reaches
10 mW (the combustion peak is not Petroleum Fraction Abbreviation Sample weight Onset [°C]
[mg]
of interest).
• Atmosphere: Oxygen at 3 MPa, no Light Cycle LCGO 4.805 173.3
purge gas. Kerosine 6.151 203.5
Vacuum VGO1 6.391 210.4
Interpretation Vacuum VGO2 6.016 210.6
The high pressure of oxygen prevents
Vacuum VGO3 6.635 214.6
any premature vaporization so that, on
Light Vacuum LVGO 5.578 224.9
heating, a deflection in the DSC curve is
caused only by a change in the specific Light LGO 5.300 228.7

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 13

 4. D
 etermination of the Noack evaporation loss
of lubricants by TGA
Application

4.1 Abstract viscosity, evaporation loss and shear sta- Acronyms

For quality and environmental reasons, bility. Performance behavior is also test- API: American Petroleum Institute
lubricants for engines and other appli- ed in engine tests. ASTM: A STM International, originally
cations must only exhibit a low evapo- known as the American Society
ration rate. The loss of volatile compo- This includes wear protection and clean- for Testing and Materials
nents from an oil increases its viscosity liness as well as the influence on fuel DIN: Deutsches Institut für Nor-
and leads to increased oil consumption, consumption and the changes in the en- mung (German Institute for
coking and wear. gine oil during operation due to viscosity Standardization)
changes (thickening). The classification ILSAC: International Lubricant
The Noack method is a widely used is provided by organizations such as IL- Standardization and Approval
standard test method for measuring the SAC, API or SAE (see the table of acro- Committee
evaporation loss from lubricating oils. nyms). JPI: Petroleum Association of Japan
According to the ILSAC GF-3 and API-SL SAE: SAE International, Society of
specifications the evaporation loss must One of the commonly used specifications Automotive Engineers
not be greater than 15%. is the evaporation loss. The low molecu-
lar mass constituents of an engine oil,
The ASTM standard test method D6375 which consists of fractions of differ- to combine the advantages of the gas
for the determination of the evapora- ent hydrocarbons with different chain chromatographic method [6] with the
tion loss of lubricating oils by the No- lengths and molecular masses, can evap- realistic conditions of the traditional No-
ack method [1] uses thermogravimetric orate under increased thermal stress. ack test. The method is quicker and more
analysis, TGA. reliable than both and can be performed
This usually leads to an increase in the with less sample material.
This method yields the same results as viscosity of the lubricant. At the same
other standard test methods (e.g. ASTM time, the solubility of the additives in the According to the ASTM D6375 method, a
D5800 [2], DIN 51581-1 [3], JPI-5S-41-93 base oil is affected. The evaporation is sample is heated rapidly in a crucible to
[4]). important for all lubricant groups (e.g. 249 °C and held isothermally for 30 min
also for synthetic oils) if they are used at at this temperature during which time
This article describes how the Noack higher temperatures. the TGA curve is recorded. The Noack
evaporation loss is determined in com- evaporation loss is the loss in mass up to
parison to a reference oil sample using For example with engine oils, evapora- the Noack reference time.
TGA. tion losses can occur through high tem-
peratures at the piston rings and else- This time is determined beforehand
4.2 Introduction where. These losses lead to undesirable under the same experimental condi-
The increase in the usable lifetime of oil thickening and increased oil con- tions with a Noack reference oil. In this
lubricants coupled with faster oil circu- sumption. method, it is important that the sample
lation rates, longer oil change intervals temperature quickly rises to a value be-
and lower lubricant consumption means 4.3 The Noack evaporation loss tween 247 and 249 °C but does not over-
that lubricants are subjected to greater test according to ASTM D6375 shoot.
stress. The Noack test to quantitatively deter-
mine the evaporation loss of oils under To simulate the traditional Noack meth-
Higher temperatures coupled with small- standard conditions was introduced od, a sample is usually heated to 220 °C at
er oil volumes and higher performance many years ago. 100 K/min and then to 249 °C at 10 K/min.
lead to a constant increase in the de- The sample mass (ms) to be used is de-
mands placed on the performance and For example, the DIN 51581 [3] test termined from the internal diameter
quality of the lubricants. To ensure that method measures the evaporation loss of the crucible (d) using the following
the lubricants are properly used, they over a period of one hour at 250 °C under equation:
must be properly specified and classified. vacuum (2 mbar).
ms = 350d3
The specifications describe the physi- The ASTM D6375 standard thermogravi-
cal properties of engine oils such as the metric test method was developed [5] where d is in cm and ms in mg.

14 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

4.4 Performing a Noack test furnace should be heated to 1000 °C evaporation loss by thermogravimetry
The Noack evaporation test (also referred without a crucible and held isothermally is therefore an important step in the
to as Noack volatility) was performed at this temperature for about 5 min. The qualification of lubricant. The METTLER
under the following conditions using a air gas flow is left at about 80 mL/min. TOLEDO TGA system with sample robot
METTLER TOLEDO TGA instrument: and automated evaluation provides high
• Crucible: 100 µL aluminum without 4.5 Conclusions sample throughput and rapid pass/fail
lid (internal diameter 0.56 cm) Over the past years, the demands placed assessment of the oil in question.
• Sample mass: 61 ± 3 mg on lubricants in many application areas
• Purge and protective gas: total have changed significantly. 4.6 References
80 mL/min air [1] ASTM D6375 Standard Test Method
for Evaporation Loss of Lubricating
• Noack reference oil: W4520001 with Thermal oxidation stability, low tendency Oils by Thermogravimetric Analyzer
10.93% mass loss up to the Noack to evaporate and their influence on our (TGA) Noack Method.
time; supplier: Walter Herzog GmbH natural and working environments have [2] ASTM D-5800 Test Method for
Evaporation Loss of Lubricating Oils
• Temperature program: 50 to 220 °C become very important. by the Noack Method.
at 100 K/min, followed by heating to [3] DIN 51581-1, 2003-02 Determination
of the evaporative loss of petroleum
249 °C at 10 K/min and isothermal The innovative development of modern
products by the Noack method –
at 249 °C. To achieve the above lubricants and their proper application Part 1 (original in German).
condition, the τlag parameter for this have far-reaching economical conse- [4] JPI-5S-41-93 Determination of
Evap-oration Loss of Engine Oils
crucible was adjusted to zero. quences. Lubricants (base liquid and (Noack Method).
additives) that have been optimized for [5] E. F. de Paz, C. B. Sneyd – The Thermo-
The sample (test oil) used was a synthetic the different tasks, for example with low gravimetric Noack Test: a Precise,
Safe and Fast Method for Measuring
5W40 engine oil. evaporation losses, Lubricant Volatility, Subjects in Engine
• save energy Oil Rheology and Tribology, SP1209,
Figure 1 shows the TGA curve (black • reduce service intervals International Fall Fuel and Lubricants
Meeting, San Antonio 1996, available
line) of the reference oil. The Noack ref- • minimize wear also as SAE Technical Papers,
erence time is 17.56 min; this is the time • increase engine service life Document Number: 962035.
at which the certified loss of 10.93% is • increase oil change intervals (lifetime) [6] DIN 51581-2, 1997-05 Determination
of the evaporation loss of petroleum
reached (see the black arrows). and result in considerable economi- products by gas chromatography –
cal savings. The determination of the Part 2 (original in German).
The TGA curve of the sample is shown as
Figure 1.
the red curve. After 17.56 min the loss is Determination of the
read off as 7.80% (see red arrows): the evaporation loss
(Noack volatility)
synthetic 5W40 engine oil thus has a from a synthetic
Noack evaporation loss of 7.8%. This is engine oil using
often given in the oil specification simply the TGA Noack test
method.
as NOACK 7.8%. Black continuous
curve: reference oil;
According to ASTM D6375, the repeat- red curve: synthetic
5W40 engine oil;
ability with 8% loss is about 1% for two black dashed curve:
determinations by the same laboratory, sample temperature.
and the reproducibility is about 1.4% for two
determinations by different laboratories.

According to ASTM D6375, the TGA fur-

nace must be regularly heated out. The
method recommends that this is done
after about ten determinations. The

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 15

 5. Rapid thermogravimetric analysis of coal
5.1 Introduction required for other applications. Analo- moisture is, however, still accurately
The quantitative determination of mois- gous to the analysis of coal, a standard- measured – the weight loss up to the
Application

ture, volatile compounds, chemically ized thermo­g ravimetric procedure is start is automatically measured by deter-
bound carbon, and ash content has long nowadays used to determine the content mining the starting weight on reaching
been used to determine the quality and of elastomers, thermoplastics and ther- 110 °C. This is done by measuring the
economic value of different types of coal. mosets, as well as lubricants [16, 17]. weight of the sample and the crucible
High ash content is undesirable for the during the sample preparation, which is
operation of thermal power stations Conversely, procedures developed for the in fact normal in the manual procedure
because inert material increases trans- determination of carbon black in rubber [1–10]. Buoyancy effects are compensat-
port and waste disposal costs, and also [17] are used for the analysis of brown ed in the TGA measurement by subtrac-
means that the heat exchangers have to coal, lignite and or other renewable fossil tion of a blank curve.
be cleaned more frequently. To make sure fuels [e.g. 18].
that assays can be properly compared, Besides this, oxygen is used instead of
the analysis procedures have been stan- 5.2 Speeding up the TGA air, resulting in much shorter combus-
dardized and described in many standard procedure tion times. One must of course make sure
methods [1–10]. As already indicated, the standard meth- that coal particles are not blown out of the
ods for coal analysis are laborious and crucible by the rapid generation of gas.
Quite early on, measurement routines often time-consuming.
were developed for thermogravimetric This can manifest itself for example in
instruments that enabled faster and more To determine the moisture content ex- widely differing measurement values for
automated analyses to be performed. actly, the measurement is usually started the ash content. As with all analytical
These techniques have been compared at room temperature, or slightly above. methods, it is essential that the coal sam-
with the standard manual methods This leads to long cooling times before ples are homogeneous and representative
[11–14]. the next sample can be measured, which if one, for example, wants to characterize
in turn limits the throughput of samples. one hundred tons of coal. Vaporization,
TGA (thermogravimetric analysis) The METTLER TOLEDO STAR e system degassing and combustion proceed more
is, however, also very useful for coal allows the analysis time to be reduced by rapidly when shallow, open crucibles
research, e.g. to compare combustion half compared to the ASTM E1131 stan- are used and small sample weights are
profiles or to determine the nature of vol- dard method. analyzed.
atile components with TGA-MS. Even the
lime deposits (fur) formed in hot water sys- In this case, the measurement is start- 5.3 Experimental details
tems have been investigated with TGA [15]. ed directly at 110 °C (or even higher), A METTLER TOLEDO TGA /SDTA851e
The determination of moisture, vola- which means the otherwise long cool- with the small furnace (up to 1100 °C)
tile content, soot, ash or fillers is also ing time down to 30 °C is avoided. The was used for the measurements. The sys-
tem was automated with a TSO800RO
Figure 1.
TG curves showing sample robot and a TSO800GC1 gas con-
two measurements troller for gas switching.
of each of the three
types of coal. At
10.9 min the purge The balance was purged with 40 mL/min of
gas was switched nitrogen as protective gas. During the
from nitrogen to
oxygen. One of the
measurement, first 80 mL/min of nitro-
measurements of gen, and afterward 80 mL/min of oxygen
Coal A (continuous were used as reactive gas.
curves) was used
to evaluate the
weight loss steps. The following samples were used as
The uppermost examples for coal analysis:
horizontal curve is
a measurement 1. Coal A with a particle size of 65 µm to
without a sample 90 µm,
(i.e. a blank curve). 2. Coal B with a particle size up to
1 mm,
3. Coke, finely powered, manufactured
from petroleum oil.

16 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

 Table 1.
Coal type Method Moisture Volaitle content Coal Ash Residue
Comparison of
Coal A Fast (M) 2.80 31.36 62.45 3.38 3.48 the contents
(SD) 0.01 0.05 0.03 0.06 0.02 measured using
the fast method
ASTM 2.53 31.59 62.46 3.21 with the values
Coal B Fast (M) 4.69 28.45 47.80 19.02 19.11 according to
(SD) 0.05 0.33 0.66 1.00 1.02 the ASTM E1131
standard. In each
ASTM 4.27 28.66 47.62 18.89 case, the mean
Coke Fast (M) 0.00 0.48 99.38 0.13 0.10 values (M) and
the corresponding
(SD) 0.00 0.01 0.02 0.02 0.00 standard deviations
ASTM 0.01 0.48 99.38 0.17 (SD) are calculated
from three measure-
ments. The residue
is determined by
Unless otherwise specified, alumina cru- to the sample weight, i.e. the total weight cluding evaluation takes approximately manually reweigh-
cibles (30 µl) were filled with about 20 mg is determined directly after weighing- 36 minutes; of this, about 15 minutes is ing when the
of coal powder. During weighing-in, the in. After insertion of the sample by the for the actual measurement. The assay measurement is
completed as
crucible weight and the sample weight were sample robot and the equilibration of the according to ASTM E1131 takes about 60 specified in the
noted and entered in the software program. temperature at 110 °C (automatic “set- minutes because somewhat longer iso- standards
tling”), the total weight is determined at thermal times are necessary and because [2 and 6]. All the
values refer to
Immediately after weighing-in, the this instant with the first measurement an additional cooling time of 10 minutes the initial sample
special aluminum lid was placed on the point. is needed to reach 30 °C. weight.
crucible to prevent contamination of the
sample and any large change in moisture The difference between this and the As Table 1 shows, the corresponding
content during the waiting time. The lid sample weight before the measurement mean values for both methods are very
is removed by the sample robot during yields the dry content (or the amount of close. The differences are less than the
measurement. moisture evolved). The starting weight is reproducibility of >0.3% given in the
given by the Signal Value (Fig. 1) at the manual methods [5–7].
At the end of the measurement, the mea- beginning of the normalized TGA curve.
suring cell automatically cools down This means that drying or moisture up- It is noticeable that there is a tendency for
to the starting temperature. The blank take during the waiting time up to the the fast method to yield higher moisture
curve is automatically subtracted and the actual measurement has no influence on values and lower volatile contents. This
evaluation macro-program calculates the original moisture content. can be explained by a loss due to drying
the weight loss immediately after the of about 0.4% with the ASTM method [16]
measurement is finished. However, especially with very moist coal before the actual measurement. This can
samples, the use of an aluminum lid is only be prevented if the weighed-in sam-
5.4 Results recommended as described in the experi- ple is immediately measured.
Fast method mental section. In the first 30 seconds,
Figure 1 shows the TGA curves of the coal the TG curves are practically horizontal The comparatively high standard devia-
analyses performed with the fast method with a loss of less than 3 µg (this corre- tions with Coal B are also worth noting.
described above. sponds to 0.01%), proving that drying is This is also apparent in the manually
complete at the beginning of the mea- weighed residues.
The measurement conditions were cho- surement. The curves are also horizontal
sen as follows: Under nitrogen: 110 °C for at the end of the isothermal phase at 900 Measurements in crucibles with alumina
0.5 min, then heating at 100 K/min to °C because degassing and combustion lids (with a hole) and in taller crucibles
900 °C and held at this temperature for have finished. (70 µl) showed that the differences can-
2.5 minutes. not however be explained through too
The measured values are summarized rapid combustion with “firework-like”
The system was then purged with oxygen in Table 1. The reproducibility is shown ejection of material (“sparkling coal”).
for the following 4 minutes (furnace at- by the standard deviation and how well Better grinding and homogenization of
mosphere is made up of 80 mL/min O2 the measurement curves coincide. The the sample however reduced the standard
and 40 mL/min N2). The blank curve was slight differences in combustion rates are deviations by a factor three. The reason
measured in the same way before the mea- caused by small differences in the sample is the inhomogeneity of this coal sample.
surement series with an empty crucible. weights, e.g. with Coal B 19.801 mg as This underlines the necessity of good
For all measurements (including the opposed to 21.547 mg (see Fig. 1). sample preparation. The comparison of
blank curve), the weight of the crucible the two coal types A and B show this very
is also entered in the software in addition In routine operation, the analysis in- clearly.

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 17

 5.5 Conclusions 5.6 References 11. Richard L. Fyans,
1. DIN 51718 Analysis of solid fuels – “Rapid Characteriza­tion of Coal by
The assay of moisture, volatile content, Thermogravimetric and Scanning
determination of water content and
carbon and ash in coal and coke using analysis moisture (in German). Calorimetric Analysis”, Presentation
TGA is a standardized routine method. 2. DIN 51719 Analysis of solid fuels at the 28th Pittsburgh Conference in
– determination of ash content. (in Cleveland, Ohio, March (1977).
The time required for the method can be 12. John W. Cumming, Joseph McLaughlin,
Application

German)
shortened by up to 50% without affect- 3. DIN 51720 Analysis of solid fuels – “The thermogravimetric behavior
of coal”, Thermochimica Acta, 57
ing the accuracy by using the METTLER determination of the content of volati-
le components. (in German) (1982) 253-272.
TOLEDO STARe system and adapting the 13. F. S. Sadek, A. Y. Herrell, “Proximate
4. ASTM D3172 Standard Practice for
temperature program. In automated op- Proximate Analysis of Coal and Coke. analysis of solid fossil fuels by
5. ASTM D3173 Standard Test Method thermo­gravimetry”, American
eration, this allows at least 35 analyses to Laboratory, March (1984) 75–78.
for Moisture in the Analysis Sample
be performed per day. of Coal and Coke. 14. Danny E. Larkin, “The development
6. ASTM D3174 Standard Test Method of a standard method”, ASTM STP
for Ash in the Analysis Sample of 997 “Compositional analysis by
The TGA curves show whether degas- thermo­gravimetry”, C.M. Earnest, Ed.,
Coal and Coke.
sing and combustion are complete. This 7. ASTM D3175 Standard Test Method American Society for Testing and Ma-
is necessary in order to optimize the for Volatile Matter in the Analysis terials, Philadelphia (1988) 28–37.
Sample of Coal and Coke. 15. Paul Baur, “Thermogravimetry
method and for the control of routine speeds up proximate analysis of
8. ISO 11722 as well as BS 1016-104.1
measurements. Thermogravimetry, in Methods for analysis and testing of coal”, Power, March (1983) 91–93.
addition, allows drying and combustion coal and coke. Proximate analysis. 16. ASTM E1131 Standard Test Method
Determination of moisture content of for Compositional Analysis by
behavior to be investigated and different Thermo­gravimetry.
the general analysis test sample.
coal types to be characterized. 9. ISO 562 as well as BS 1016-104.3 17. ISO 9924 Rubber and rubber products.
Methods for analysis and testing of Determination of the composition
coal and coke. Proximate analysis. of vulcanizates and uncured com-
Inhomogeneous samples must be well pounds by thermogravimetry.
Determination of volatile matter
ground to achieve a high degree of repro­ content. 18. M.C. Mayoral, et. al., “Different
10. ISO 1171 as well as BS 1016-104.4 approaches to proximate analysis
ducibility. If necessary, larger amounts by thermogravimetry analysis”,
Methods for analysis and testing of
of sample can be measured thermogravi­ coal and coke. Proximate analysis. Thermo­chimica Acta, 370 (2001)
metrically in 900 µl crucibles. Determination of ash content. 91–97.

18 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

6. D
 etermination of oxidation stability by
pressure-dependent OIT measurements
6.1 Introduction sure and flow regulator. 6.3 Pressure dependence of
The determination of the oxidation The gas flow was 40 mL/min and the OIT
induction time (OIT) is a method oxygen pressure set to different values Experimental results
frequently used to estimate the oxidation in the range 0.1 MPa to 10 MPa (1 to If the OIT is measured at constant reac-
stability of materials. 100 bar). Since gas switching in pressure tion temperature but at different oxygen
measurements is not always so easy, the pressures, one sees that the OIT decreases
This is done by first heating the sample oxygen pressure was set at room temper- with increasing pressure.
to a sufficiently high temperature under ature and the sample heated rapidly at
inert conditions (purging the DSC cell 50 K/min to the reaction temperature. This is illustrated in Figure 2 using min-
with nitrogen). After a brief thermal eral oil as an example. At an oxygen
equilibration time, the purge gas is The chemiluminescence measurements pressure of 0.5 MPa, the OIT is more than
switched from nitrogen to oxygen. If the were performed in an HP DSC827e using 120 min. In contrast, at 6.8 MPa the
temperature has been chosen correctly a CCD camera and the relevant accesso- OIT is less than 40 min. Besides shorter
there is no immediate change in the DSC ries (see UserCom 20). measurement times, the measurement
signal.
Figure 1.
OIT measurements
However, after a certain induction performed in a
conventional DSC at
time, the exothermic oxidation reaction
atmospheric pres-
begins. The OIT is the time at which sure. The material
the onset of the oxidation reaction oc- originated from a
piece of PE tubing.
curs measured from the point when the The reaction tem-
gas is switched from nitrogen to oxy- perature is 210 °C.
gen. The OIT depends very much on the The stressed (i.e.
damaged) sample
current chemical state of a material. material originated
For example, chemical aging causes a from the inside
significant decrease in the measured of the tubing, the
unstressed sample
OIT. from the outside.

An example of this is shown in Figure 1.

This shows OIT measurements performed
at 210 °C on polyethylene (PE) tubing
that had been used to transport aqueous
solutions of chemicals. Two samples, one
Figure 2.
from the outside and the other from the OIT measurements
inside were measured. The inside of the of samples of a
mineral oil (A) at
tubing had of course been in direct con-
different oxygen
tact with the solutions. pressures. The re-
action temperature
is 195 °C.
The OIT of the material from the outside
of the tubing is significantly longer than
that from the inside. The OIT therefore
provides information on the extent of
damage of materials.

6.2 Experimental details

The measurements described in this
study on the pressure dependence of OIT
were performed using the new high pres-
sure DSC (HP DSC827e). The oxygen pres-
sure and gas flow were set using a pres-

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 19

 reproducibility also improves at higher log tOIT = log A –q log p (2) 6.4 Interpretation of the results
pressure. In this section, possible reasons for the
The factor A is a measure of the rate of the power law given in eq (1) will be consid-
Figure 3a shows the pressure dependence oxidation reaction and the exponent q is ered without going into a detailed discus-
of the OIT of two different mineral oils, the slope. The q values obtained for the sion about the reaction kinetics.
Application

A and B at a reaction temperature of two oils are 0.62 (oil B) and 0.45 (oil A).
195 °C. The OIT of both oils decreases First, we assume that the presence of oxy-
with increasing pressure, whereby the The general validity of the potential law gen is necessary for the oxidation reac-
OIT values of oil B are always greater and the meaning of the exponent q must tion. The oxygen enters into the sample
than those of oil A. first be considered before the results can through diffusion from the atmosphere.
be assessed in more depth. This was done This process depends on the difference
In Figure 3b, the values have been plotted by performing OIT measurements on a in concentration, the size and the nature
in a double logarithmic presentation in number of different polymers. Some of the of the surface of the sample and the tem-
order to analyze the relationship between results obtained are shown in Figure 4. perature-dependent diffusion constants.
OIT and oxygen pressure more easily.
Different reaction temperatures were At the same time, the effective oxygen con-
In this presentation, the measured points chosen for each polymer so that the OIT centration is reduced through reaction of
lie on straight lines. values were all in a convenient time the oxygen with stabilizers or the sample.
frame (between 20 and 300 min). This
It follows that the relationship between facilitated the analysis of the pressure de- Diffusion (production) and reactions
OIT (tOIT) and the oxygen pressure (p) pendence of OIT for the materials tested. (consumption) are competing processes
can be described by a power law: that determine the oxygen concentration.
The diagram shows that in double loga- A relationship can therefore be derived
–q
tOIT = A p  (1) rithmic presentation the relationship is for the relative oxygen concentration in
linear for all the polymers investigated. the sample as a function of time.
Taking logarithms of both sides of the The exponent q here lies between 0.6
equation shows the linear relationship in (linear low-density polyethylene, LLDPE), Similar to the situation in crystallization,
the double logarithmic presentation: and 0.3 (polypropylene, PP). it can be concluded that the oxidation re-
action begins on nucleation points such
Figures 3a (left)
and 3b (right). as surface damage (e.g. microfissures) or
Linear plot of the inhomogeneities (internal stresses or con-
OIT of two synthetic
centration differences of inhibitors) and
motor oils as a
function of oxygen then spreads from these reaction nuclei
pressure at 195 °C into all parts of the sample.
(left). Double loga-
rithmic plot (right);
the slope of the If, in addition, the pressure dependence
straight lines is –q. of the gas concentration in the sample is
taken into account, the power law eq (1)
Figure 4.
Dependence of for the OIT can be derived from kinetic
the OIT on oxygen considerations. The exponent, q, depends
pressure for differ-
to a large extent on the mechanism for
ent polymers. The
reaction tempera- the formation of nuclei for the oxidation
ture is given in the reaction.
diagram. The slope
–q corresponds
to the exponent in To confirm the assumption that the oxi-
eq. (1). dation reaction spreads out from nuclei
requires a method that can measure
the reaction locally at high resolution.
Chemiluminescence is an excellent tech-
nique for doing this.

During the oxidation reaction, excited

molecules are formed that emit light of
low intensity when they return to the
ground state.

20 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

The light emission can be measured is proportional to the oxygen pressure. This is shown schematically with the rel-
using the DSC-Chemiluminescence System. This means that the variation of OIT evant slopes in Figure 6.
This consists of the HP DSC827e equipped with pressure depends only on the oxygen
with the chemiluminescence accessory concentration in the sample. If the slope, A large value of the slope means a high
and a high-sensitivity CCD camera. In this q, equals 0, the oxygen concentration is degree of inherent oxidation stability,
case, the system enabled both intensity of no importance. which can be reduced through mechani-
curves of the emitted light and sequences cal or chemical stress. A material with a
of emission images to be recorded. In this case, the sample is damaged to smaller q value thus contains a higher
such an extent that hardly external oxy- concentration of radicals.
Figure 5 shows the intensity curve and gen is necessary for the oxidation. In
emission images at different times for a material like this, many radicals are In the example in Figure 4, LLDPE was
polypropylene (PP) at 130 °C. Evalu- present. the most stable. The PP samples had
ation of the intensity curve yields an been stored for several years and were
OIT value of 230 min. In the emission From kinetic considerations at the be- more prone to oxidative stress.
image at 205 min, a region can be seen ginning of the oxidative decomposition
at the lower edge of the sample where and taking into account the oxygen dif- For comparison purposes, Figure 4 also
light emission begins. fusion and nucleation of the reaction, includes a sample of LLDPE stabilized
q values between 0.75 and 0.2 are to be with an antioxidant. The antioxidant
This is the initial oxidation reaction. expected. hardly has any effect on the slope, q.
In the image at 285 min, three main
Figure 5.
regions of the reaction are apparent. The Results of a
regions are highlighted by arrows in the chemiluminescence
box. measurement
of PP at 130 °C.

With increasing reaction time, the

regions spread until finally the entire
sample emits light. This result proves
that the reaction spreads from localized
nuclei.

The sample under investigation was a

thin film. The orientation of the nuclei
indicates that they originated as a result
of the film production process.

The chemiluminescence measurements

therefore confirm the assumptions lead-
ing to the power law, eq (1).
Figure 6.
Schematic diagram
6.5 Assessment of experimental of the possible
slopes in double
results logarithmic pre-
In the previous section, it was shown that sentation (log t OIT
the pressure dependence of the OIT can versus log p) with
the corresponding
be traced back to nucleation processes in exponents from
the oxidation reaction, and that the nucle- eq (1). The practi-
ation is a result of local inhomogeneities cally relevant region
for q lies between
in the sample and diffusion. 0.75 and 0.2.

These considerations lead to the power

law tOIT = A p –q, where q is the slope in a
double logarithmic plot (log tOIT versus
log p) is.

Theoretically, the slope must lie between

0 and 1. If the slope, q, equals 1, the OIT

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 21

 The reaction rate is lowered, but the in- cess before the reaction. This means that It is important to remember when per-
herent tendency to oxidation remains al- the oxidation does not begin through- forming OIT measurements that the ge-
most unchanged. out the entire sample, but just in a few ometry of samples plays an important role
isolated regions in the sample where because of its influence on diffusion pro-
6.6 Conclusions conditions are favorable for diffusion cesses. For this reason, it is best to use very
Application

The pressure dependence of OIT obeys (e.g. microfissures) or where there is a thin samples with defined surface areas.
a power law. The exponent q is mainly greater concentration of radicals. This The author would like to thank his col-
influenced through oxygen diffusion nucleation mechanism in the oxidation leagues Dr. R. Riesen, M. Zappa and
processes and radical formation in the reaction can be measured by chemilu- Dr. M. Schubnell for valuable help and
material under test. Materials with small minescence. discussions.
exponents tend to oxidize more readily
than those with large exponents. The pressure dependence of OIT provides
additional information on the stability,
The power law dependence of OIT on previous damage, and inherent tendency
pressure presupposes a nucleation pro- for oxidation of materials.

22 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

7. S
 oot analysis study of engine oil samples
7.1 Purpose of analysis mass loss step after switching gas to 7.4 Conclusions
Determination of performance charac- oxygen above 650 °C and according to The soot content in the engine oil may cause
teristics of engine oil samples by % soot ASTM D5967, this mass loss is termed sludge formation, wear and filter plugging.
content/concentration. soot. In the DSC curve this is seen as a
highly exothermic reaction. Soot accumulates in the engine oil
Sample of a diesel engine as a result of engine
Five different engine oil samples on a The white colored residue or ash ob- timing and combustion and thus affects
support, abbreviated by name as: served at the end of analysis is generally its performance.
1] (A-761) composed of traces of metal oxides, non-
2] (B-502) reactive oil additives or crusted matter Note that, during soot analysis one has
3] (C-532) (dust). For sample (A-761), no soot was to make sure that no oxygen is present in
4] (D-636) present. TGA furnace when a new measurement
5] (E-663) is started.
7.3 Evaluations
Conditions The numerical values of each mass loss This can be achieved by keeping the
• Measuring cell: TGA/SDTA851e step are given in the diagram and in the sample for 25 min at ambient tempera-
(1600 °C) with GC100 table in the evaluations. % Soot content ture in the TGA furnace before the mea-
• Pretreatment: All samples shaken & Residue content of all five different surement and exposing it to a nitrogen
well before analysis samples was as follows: gas flow of 50 mL/min.
• Sample preparation: Max. 20 mg of
sample in 70 µL Alox crucible Sample % Soot content Residue (ash content)
• Measurement Temp. program with (650 °C–750 °C)
6 segments (ASTM:D5967-04) Gulf oil (A-761) 0.05 % (negligible) 1.01 %
1) 1 min Isothermal at 50 °C, N2 Gulf oil (B-502) 1.23 % 1.14 %
2) 50 to 550 °C, at 100 K/min, N2
Gulf oil (C-532) 4.46 % 1.07 %
3) 1 min Isothermal at 550 °C, N2
Gulf oil (D-636) 6.65 % 1.15 %
4) 550 to 650 °C, at 20 K/min, N2
Gulf oil (E-663) 10.48 % 1.07 %
5) 650 to 750 °C, at 20 K/min, O2
6) 5 min Isothermal at 750 °C, O2
Figure 1.
• Atmosphere: 40 mL N2/min and Determination of
40 mL O2/min performance char-
acteristics of engine
oil samples by %
7.2 Interpretation soot content/con-
The test procedure applied in this re- centration.
port is part of ASTM D5967 standard test
method commonly known as Mack T-8.
It characterizes the performance char-
acteristic of diesel engine oils by its soot
concentration.

The first two mass loss steps from

50 ˚C to 650 ˚C (Figure 1) are due
to loss of volatile oil components in
a nitrogen atmosphere. All samples
were black in color except for (A-
761). These black samples exhibited a

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 23

 8. Melting behavior of oil/wax mixtures
8.1 Purpose of analysis Conditions 80 °C to −60 °C at 3 K/min
Comparison of the melting behavior of • Measuring cell: DSC822e with −60 °C isothermal, 5 min
Application

various oil/wax mixtures; reproducibility Intracooler −60 to –80 °C at 5 K/min

of the measurements. • Pan: 20 µl aluminum 50 mL/min nitrogen
• Sample preparation: Samples have
Sample been premelted and carefully stirred 8.2 Evaluation
10 samples with different oil/wax contents • Measurement: From the heating curves the total melt-
(7 measured, 4 are shown below) 80 °C isothermal, 5 min ing enthalpy was calculated. In general,
oils have a lower melting enthalpy than
waxes.

Therefore, the samples with a higher oil

content should exhibit a lower overall
melting enthalpy. Plotting the oil content
as a function of the overall melting en-
thalpy delivers a calibration curve. In the
figure below, this is shown for the mea-
sured samples.

The figure shows that the relationship

between the oil content and the overall
Figure 1.
Melting behavior of
melting enthalpy can be described in the
samples with available concentration range reason-
different oil contents; ably well by a straight line. This allows
the figure shows
only the heating a simple estimate of the oil content in the
curve. two samples with unknown oil content
(see below).
Figure 2.
Heating and cooling
curves of sample 5; Note that for a reliable prediction the
this figure illustrates
melting enthalpy of the unknown sample
the excellent re-
producibility of the should be within the enthalpy range cov-
crystallization as ered by the calibration samples.
well as of the melt-
ing behavior of the
samples. 8.3 Conclusion
Based on a calibration curve that has
to be established with samples of which
the oil content is well known, the oil
content of any sample can be estimated
by simply measuring its overall melt-
ing enthalpy and using the data of the
previously established calibration curve.
All these measurements can be easily
done by a straightforward analysis of the
DSC-heating-curves.

24 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

9. S
 tandards for petrochemicals with respect
to thermal analysis
Table 1.
Standard Petrochemical Topic Measuring technique
Excerpts of
ASTM D4419 Waxes Transition temperatures DSC standards for
ASTM D6186 Lubricating oils, Oxidation induction time (OIT) DSC, HP DSC petrochemicals
ASTM D5483 lubricating greases
ASTM D6375 Lubricating oils Evaporation loss (Noack) TGA
IP 389 Fuels Wax appearance temperature DSC
(WAT)
ASTM E1131 General Compositional analysis TGA
ASTM E2009 Hydrocarbons Oxidation onset temperature DSC
(OOT) of hydrocarbons by DSC

The instruments of Thermal Analysis meters and pH meters meet the require-
Excellence are compliant with relevant ments of respective standards by ASTM,
standards for the analysis and character- IP and other standard owners.
ization of petrochemicals.
Balances and scales form the backbone
Analytical instruments such as poten- for various weighing tasks in the labora-
tiometric titrators, Karl Fischer titrators, tories, in production and tank farming
dropping point instruments, density as well as for vehicle weighing.

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 25

 10. For More Information
Outstanding Services
METTLER TOLEDO offers you valuable support and services to keep you informed about new developments and
Fore more Information

help you expand your knowledge and expertise, including:

News on Thermal Analysis

Informs you about new products, applications and events.
www.mt.com/ta-news
www.mt.com/ta-app

Handbooks
Written for thermal analysis users with background information, theory and practice, useful tables of material
properties and many interesting applications.
www.mt.com/ta-handbooks

Tutorial
The Tutorial Kit handbook with twenty-two well-chosen application examples and the corresponding test
substances provides an excellent introduction to thermal analysis techniques and is ideal for self-study.

Title Order Number

Tutorial Kit (handbook only) 30281946
Tutorial Kit (handbook and samples) 30249170

www.mt.com/ta-handbooks

Videos
Our technical videos explain complex issues concerning thermal analysis instrumentation and the STARe
software – whether it’s sample preparation, installation, creating experiments or evaluating measurement results.
www.mt.com/ta-videos

UserCom
Our popular, biannual technical customer magazine, where users and specialists publish applications from
different fields.
www.mt.com/ta-usercoms

Applications
If you have a specific application question, you may find the answer in the application database.
www.mt.com/ta-applications

Webinars
We offer web-based seminars (webinars) on different topics. After the presentation, you will have the opportunity
to discuss any points of interest with specialists or with other participants.
www.mt.com/ta-webinars (Live Webinars)
www.mt.com/ta-ondemand (On Demand Webinars)

Training
Classroom training is still one of the most effective ways to learn.
Our User Training Courses will help you get the most out of your equipment. We offer a variety of one-day theory
and hands-on courses aimed at familiarizing you with our thermal analysis systems and their applications.
www.mt.com/ta-training (Classroom)
www.mt.com/ta-etraining (Web-based)

26 Thermal Analysis of Petrochemicals METTLER TOLEDO Application Examples

METTLER TOLEDO Application Examples Thermal Analysis of Petrochemicals 27
Overview of METTLER TOLEDO
Thermal Analysis Application Handbooks
The following application handbooks are available can be purchased:
www.mt.com/ta-handbooks

Introductory handbooks Language Order number Details

Thermal Analysis in Practice English 51725244
Volume 1
Fundamental Aspects (327 pages)
Thermal Analysis in Practice English 30306885
Volume 2 German 30315767
Tips and Hints (48 pages) French 30315768
Italian 30353357
Russian 30315769
Japanese 30315900
Korean 30315901
Thermal Analysis in Practice English 30281946 Handbook and
Volume 3 Tutorial samples
Tutorial Examples (86 pages) 30249170
Validation in Thermal Analysis English 51725141
A Guide (232 pages)

Applications handbooks Language Order number Details

Thermal Analysis of Elastomers English 51725061 Volumes 1 and 2
Volumes 1 and 2 51725057 Volume 1
Collected Applications (275 pages) 51725058 Volume 2
Thermal Analysis of Thermoplastics English 51725002
Collected Applications (150 pages)
Thermal Analysis of Thermosets English 51725069 Volumes 1 and 2
Volumes 1 and 2 51725067 Volume 1
Collected Applications (315 pages) 51725068 Volume 2
Thermal Analysis of Pharmaceuticals English 51725006
Collected Applications (100 pages)
Thermal Analysis of Food English 51725004
Collected Applications (65 pages)
Evolved Gas Analysis English 51725056
Selected Applications (65 pages)
Thermal Analysis of Polymers English 30076210
Selected Applications (40 pages) German 30076211
French 30076212
Russian 30076213
Japanese 30076214
Korean 30076215

www.mt.com/ta-news
For more information

Mettler-Toledo GmbH, Analytical

CH-8603 Schwerzenbach, Switzerland
Tel. +41 44 806 77 11
Fax +41 44 806 72 60

Subject to technical changes

© 03/2017 Mettler-Toledo GmbH, 30403900
Marketing MatChar / MarCom Analytical