Methods

Introduction

Elemental analysis of carbon (C), hydrogen (H), and nitrogen (N) is one of the most fundamental analytical techniques used to characterize and confirm the elemental composition of organic compounds. For many organic substances, CHN analysis provides the primary, and often the only, evidence of sample composition and purity. Since numerous organic compounds contain only carbon, hydrogen, nitrogen, and oxygen, the determination of C, H, and N is generally sufficient for routine characterization. Although oxygen can also be determined, it is typically analyzed separately and less frequently than CHN.

Historically, CHN analysis was performed using a PerkinElmer 2400 CHN Elemental Analyzer, which was introduced in 1989. Prior to its adoption, elemental analyses were carried out using the PerkinElmer 240 CHN Elemental Analyzer, the predecessor to the 2400 model. Following its replacement, the earlier instrument was repurposed for the determination of oxygen. These instruments are no longer in use for our current analytical services.

Measuring Principle

The sample is weighed into a tin capsule. Typically, 2-3 mg of organic material is required, although sample masses of up to 10 mg may be used for inorganic materials with low carbon content. After being carefully folded to completely enclose the sample, the tin capsule is placed in the autosampler.

The capsule is dropped into the combustion reactor, where an excess of oxygen has been introduced beforehand. At approximately 990 ^oC, the sample undergoes rapid combustion (mineralization). Under these conditions, carbon monoxide may initially be formed despite the excess oxygen.

Complete oxidation is achieved as the combustion gases pass over a tungsten trioxide (WO₃) catalyst. The resulting gas mixture consists primarily of carbon dioxide (CO₂), water (H₂O), and nitrogen oxides (NO_x), together with residual excess oxygen.

The gas mixture then passes through a silica tube packed with copper granules maintained at approximately 500 ^oC. In this reduction zone, the remaining oxygen is removed, and the nitrogen oxides are reduced to nitrogen gas (N₂). The resulting gas stream therefore contains the analytically relevant components CO₂, H₂O, and N₂. Any sulfur dioxide (SO₂) or hydrogen halides that may be present are removed by appropriate absorption traps.

High-purity helium (99.999%, Grade 5.0) is used as the carrier gas. The purified gas mixture is brought to a defined pressure and volume before entering the gas chromatographic separation system. Separation of the components is performed by zone chromatography, in which the individual gases are detected as a series of step-shaped signals. The height of each step is directly proportional to the amount of the corresponding component.

Blank measurements are obtained using empty tin capsules to account for background contributions.

Calibration is performed by analyzing certified reference materials supplied by the instrument manufacturer. The measured responses are used to establish calibration factors for the quantitative determination of carbon, hydrogen, and nitrogen.

Working Range

The combustion and detection process enables the determination of all analyte elements present in the sample. In addition to purely organic compounds, a wide range of metal–organic compounds and even inorganic materials, such as carbides and nitrides, have been successfully analyzed.

For sample amounts of 2-3 mg, the detection limit for carbon and nitrogen was determined to be approximately 0.05 wt% (500 ppm). At this concentration level, the measurement uncertainty is approximately 0.02 wt%.

According to the instrument manufacturer, the uncertainty of the analyzer in the medium concentration range remains below 0.3 wt%, fulfilling the requirements commonly applied by scientific journals for confirming the expected elemental composition of a compound. For samples with very high carbon content, slightly higher deviations were observed. However, the uncertainty remained within approximately 0.5 wt%.

Problems and Interferences

The analysis of oily or liquid samples is not possible using the standard thin-walled tin capsules, as these materials cannot be handled reliably in this form. For such samples, alumina crucibles with lids are available. These crucibles are sealed by cold welding to prevent sample loss due to leakage and to minimize evaporation during the combustion process. However, the enclosed air volume leads to a significant increase in the nitrogen blank value. Consequently, the detection limit for nitrogen in liquid samples is increased to approximately 0.1–0.2 wt%.

Highly viscous or glass-like materials may also be unsuitable for elemental analysis using the standard procedure, as complete combustion and homogeneous sample handling cannot always be ensured.

It has long been recognized that phosphorus can interfere with the combustion process of organic materials. The formation of glassy phosphorus-containing residues, described in the literature as P₂O₅·xH₂O·yC, can prevent complete oxidation of carbon. As a result, the elemental analysis of phosphorus-containing compounds may exhibit systematic deviations in the measured carbon content exceeding the accepted tolerance limit of 0.3 wt%. This interference can be minimized by adding vanadium pentoxide (V₂O₅) as an oxidation catalyst.

Fluorine-containing compounds are converted to hydrogen fluoride (HF) during combustion. HF can react with the silica surfaces of the reaction tubes, which constitute a major part of the reaction zone. The resulting gaseous silicon fluorides (e.g., SiF₄ and related species) may introduce systematic errors. However, these effects rarely become significant compared with the 0.3 wt% tolerance limit.

The combustion of metal-containing samples may also be affected by matrix-related interferences. Many of these effects can be compensated for by modifying the analytical procedure. However, such corrections require prior knowledge of the sample composition.

Introduction

It is generally assumed that the determination of carbon, hydrogen, and nitrogen (C/H/N) is sufficient to verify the composition of an organic compound by elemental analysis. However, when heteroatoms are also present in the molecular formula, their determination should be considered. The analysis of additional elements provides stronger evidence for the validity of the proposed composition and enables the identification of deviations from the expected elemental content.

In particular, the simultaneous determination of C/H/N/S offers significant advantages, as all four elements can be analyzed within a single measurement run. Furthermore, reliable results can be obtained from relatively small sample quantities, typically in the range of 1.5-2.5 mg.

The determination of C/H/N/S has been performed using elemental analyzers over the years, starting with the EA1108 elemental analyzer, manufactured by Carlo Erba Instruments (represented by Fisons Instruments and, since 1996, part of ThermoQuest), followed by the EuroVector EA3000 elemental analyzer, and currently using the EuroVector EA3100+ elemental analyzer.

Prior to the introduction of these analyzers, sulfur determination was performed separately using offline methods. Two procedures were commonly employed. The first involved oxidative digestion of the sample in pure oxygen, followed by sulfate determination using titration. This method was suitable for most organic compounds. However, due to the harsh oxidation conditions (e.g., combustion in an electrical furnace or Schöniger flask), sulfur determination in metal-containing compounds was often unreliable or impossible.

For such samples, a reductive digestion procedure according to Zimmermann was employed. This classical wet-chemical method provided an alternative approach for sulfur determination in inorganic and metal-containing materials, although it was a more laborious and less convenient procedure.

Method

The sample material is weighed into a tin capsule. Typically, 2-3 mg of sample are required. However, reliable results can also be obtained from smaller sample quantities under suitable conditions. After the capsule has been folded and wrapped (resembling a small piece of folded tin foil), it is introduced into the autosampler.

The tin capsule containing the sample is dropped into a combustion furnace, where a defined amount of high-purity oxygen is introduced. The initial combustion step occurs at approximately 1,020 ^oC within 1-2 seconds. According to the literature, combustion of the tin capsule itself produces a temporary temperature peak of approximately 1,400 ^oC. Under these conditions, formation of carbon monoxide is likely according to the Boudouard equilibrium, despite the presence of excess oxygen.

To ensure complete oxidation, the combustion gases pass through a catalyst zone containing tungsten(VI) oxide (WO₃). The resulting oxidation products consist mainly of CO₂, H₂O, NO_x, and SO_x. Although sulfur dioxide (SO₂) is expected to be the predominant sulfur oxide species, the possible formation of sulfur trioxide (SO₃) under these conditions cannot be completely excluded. Residual oxygen also passes through the oxidation catalyst.

The gaseous products subsequently flow through a reduction zone containing elemental copper maintained at approximately 860 ^oC. At the copper surface, excess oxygen is chemically bound through the formation of copper(II) oxide (CuO), preventing oxygen from passing through the reduction zone. Nitrogen oxides (NO_x) are reduced to nitrogen (N₂), and any sulfur trioxide formed is converted to sulfur dioxide (SO₂). Consequently, the gas stream leaving this zone contains the analytically relevant species CO₂, H₂O, N₂, and SO₂, carrying the complete elemental information required for the determination of carbon, hydrogen, nitrogen, and sulfur.

High-purity helium (grade 5.0) is used as the carrier gas. Separation of the gaseous components is achieved by gas chromatography, and detection and quantification are performed using a thermal conductivity detector (TCD).

Blank measurements are obtained using empty tin capsules to account for background contributions.

Calibration is performed by analyzing standard samples of sulfanilamide, a certified reference material supplied by the instrument manufacturer. The resulting response factors are used for quantitative determination of carbon, hydrogen, nitrogen, and sulfur.

Working Range

All analyte elements constituting the sample material are detected by this method. The recovery of sulfur has been demonstrated to be close to 100%, even for sulfate-containing compounds. In addition to common organic materials, metal carbides and nitrides have also been successfully characterized.

This method is preferred over conventional C/H/N analysis when only small sample quantities are available or when accurate determination of low nitrogen concentrations is required. Since the chromatographic data are recorded using in-house developed software (S.C.A.D.A. © J. Theiner) and data processing is performed using manually adjustable evaluation algorithms, the analytical results can be assessed and verified in greater detail compared with automated "black-box" evaluation procedures.

Elemental analysis results obtained from sample quantities below 1.5 mg are generally less reliable than might be expected based solely on the method description. Nevertheless, it has been demonstrated that C/H/N/S analysis can provide meaningful results even for sample quantities as low as 100-300 µg.

For sample amounts of 2-3 mg, the detection limit for carbon, nitrogen, and sulfur was determined to be approximately 0.05 wt% (500 ppm), with an uncertainty below 0.02 wt%.

According to the instrument manufacturer, the analytical uncertainty in the medium concentration range is within 0.3 wt%. For samples with carbon contents exceeding 90 wt%, this tolerance may need to be extended to approximately 0.5 wt%. However, the accuracy of the method has been confirmed using a phenanthrene reference material, for which the elemental analysis results remained within the 0.3 wt% tolerance limit. Therefore, for extremely carbon-rich samples, deviations from the theoretical composition may often reflect the presence of minor impurities, as achieving the required purity level (approximately 99.5%) is experimentally challenging.

Limits and Interferences

Since liquid and oily samples cannot be reliably enclosed in standard tin capsules, aluminum pans with lids can be used as an alternative sample container. Although these pans are not specifically intended for use with this instrument, they have been shown to perform satisfactorily. The lid is tightly sealed by cold welding using a dedicated press to prevent sample loss due to evaporation or leakage. However, the air volume enclosed within the pan leads to a significant increase in the nitrogen blank value. Consequently, the detection limit for nitrogen in liquid samples is increased to approximately 0.1-0.2 wt%.

Highly viscous or glass-like materials may, in some cases, be unsuitable for analysis, as they cannot be reliably handled or transferred into the sample containers.

It is well known that phosphorus-containing compounds can interfere with the combustion process of organic materials. The formation of glassy phosphorus-containing residues, described in the literature as P₂O₅·xH₂O·yC, has been reported. Sulfur may also become incorporated into this condensed phase, potentially affecting the analytical results.

For phosphorus-containing samples, significant deviations in C/H/N/S analysis may occur, in some cases exceeding the accepted tolerance limit of 0.3 wt%. This interference can largely be controlled by adding vanadium pentoxide (V₂O₅) as an oxidation catalyst.

Fluorine-containing compounds may also cause analytical interferences. Under combustion conditions, fluorine is converted to hydrogen fluoride (HF), which can react with the silica components of the reaction system. The resulting effects may introduce systematic errors. However, these deviations generally remain below the established tolerance limit.

The digestion and analysis of metal-containing materials may also be affected by matrix-related interferences. Through appropriate modifications of the analytical procedure, most of these effects can be minimized or compensated for.

Introduction

The determination of oxygen content in organic samples is not routinely requested. However, our laboratory is equipped with an analytical method for oxygen determination, and we recommend this characterization as a valuable complement to C/H/N analysis, particularly for C/H compounds and for hydrophilic substances, which often contain significant amounts of water.

Oxygen analysis is performed using an EuroVector EA3000 elemental analyzer coupled with a high-temperature pyrolysis furnace (HT 1500, HekaTech). Pyrolysis is carried out at approximately 1500 ^oC under a helium atmosphere. The reactor consists of glassy carbon and is partially filled with glassy carbon fragments. A layer of granulated carbon placed above this material acts as a reducing agent.

Method

Oxygen determination using the HT1500/EA3000

The sample is weighed into a silver capsule. Typically, 1.5-2.5 mg of sample material is required. After the capsule has been closed and folded, it is placed into the autosampler.

The silver capsule containing the sample is released and falls into the pyrolysis furnace, where it is decomposed at approximately 1,480 ^oC in a pure helium atmosphere. The resulting products react with the excess carbon present in the reactor, converting the oxygen contained in the sample quantitatively into carbon monoxide (CO).

At the granulated carbon phase, oxygen is released in the form of carbon monoxide. Under the prevailing pyrolysis conditions, carbon dioxide (CO₂) and water (H₂O) are thermodynamically unstable; therefore, a quantitative conversion of oxygen to CO can be assumed.

To decompose high-melting oxides, such as SiO₂, polytetrafluoroethylene (PTFE, Teflon, ([CF₂]_n)) is added to the sample. During pyrolysis, PTFE decomposes and forms highly reactive fluorine-containing species, which promote the release of oxygen from otherwise difficult-to-decompose compounds. Reaction products such as hydrogen fluoride (HF) and fluorine (F₂) are transferred together with the product gases to the analytical system.

Besides CO, the main gaseous products detected are hydrogen (H₂) and nitrogen (N₂). Other volatile compounds, such as hydrogen sulfide (H₂S) and methane (CH₄), are removed in a cold trap maintained at -196 ^oC using liquid nitrogen. This trap also retains HF, F₂, and other aggressive fluorine-containing compounds that could damage or degrade the chromatographic column. A second purification step using a trap filled with metallic calcium provides additional removal of residual HF and F₂.

The gaseous products are separated by gas chromatography using a silica-based molecular sieve column and are detected and quantified using a thermal conductivity detector (TCD).

Blank values are determined using empty silver capsules.

Calibration is performed using certified NIST-referenced standard reference materials, specifically acetanilide, benzoic acid, and L-cystine.

Working Range

The analysis provides the total oxygen content of the sample, independent of the chemical form in which oxygen is present. 

Under normal operating conditions, the limit of quantification is approximately 0.05 wt% oxygen. The measurement uncertainty remains below 0.3 wt% for oxygen concentrations up to 50 wt%.

Problems and Interferences

The pyrolysis of metallic or metal–organic compounds may also be affected by matrix-related interferences. Under the prevailing reaction conditions, the reduction of nickel- and iron-containing compounds should theoretically be possible. However, elements such as alkaline earth metals, lanthanides, actinides, silicon, and other elements forming highly stable oxygen compounds may remain as unreduced oxides.

The formation of such stable oxide phases can lead to systematic errors in the oxygen determination. Depending on the chemical composition of the sample, these deviations may become significant and can even result in a stoichiometric underestimation of the oxygen content compared with the theoretical value.

Introduction

The determination of halogen content in chemical substances by elemental analysis is applicable to prepared organic compounds containing chlorine and/or bromine in addition to carbon, hydrogen, nitrogen, sulfur, and oxygen.

The determination of chlorine and bromine is carried out by mineralization followed by argentometric titration.

Mineralization is performed by thermal decomposition in pure oxygen, either using an electric furnace or by combustion according to the Schöniger method.

In addition, argentometric titration can be applied for the direct determination of chloride and bromide ions in dissociating or hydrolyzing compounds. In this case, the sample is extracted or dissolved in an aqueous solution, and the resulting solution is analyzed according to the procedure described below, without a preceding mineralization step.

Method

Mineralization by Schöniger Flask Combustion

Mineralization in the Electric Oven

The sample is weighed into an open platinum boat. The required sample amount is typically in the range of 2-7 mg. For substances with a low carbon content, the sample mass may be increased up to 50 mg.

The platinum boat is placed inside a fused silica tube, through which a stream of pure oxygen is passed. A small movable electric furnace is then moved along the tube, heating the sample to approximately 900–950 ^oC. Under these conditions, the substances undergo thermal decomposition.

The oxidation reactions are completed in the subsequent section of the fused silica tube, which is packed with quartz wool and contains a small platinum mesh catalyst. This section is heated by a stationary furnace and maintained at a reaction temperature of approximately 700 ^oC.

An absorption tube is connected to the fused silica reaction tube. The gaseous reaction products are collected in an alkaline absorption solution containing hydrogen peroxide at room temperature.

After completion of the mineralization, the absorption solution is rinsed from the tube using distilled water, and approximately 10-20 mL of analyte solution is collected in a titration beaker. This solution contains the halide ions Cl^- and Br^-.

Fluorine cannot be determined by this method because hydrogen fluoride (HF) immediately reacts with the large excess of silica, forming silicon tetrafluoride (SiF₄). Iodide also cannot be determined by argentometric titration because its redox potential allows the formation of elemental iodine at the furnace outlet.

In addition to the halide ions, the analyte solution contains sulfate, carbonate (which readily releases carbon dioxide from the relatively acidic solution), nitrite, and nitrate ions.

Determination by Argentometry

The analyte solution is acidified by adding two to three drops of 1:1 nitric acid to ensure an acidic medium. A silver electrode is then immersed in the solution.

The titration is performed using an automatic titrator (DL 21, Mettler-Toledo, 1989) equipped with a 1 mL/2000-step microburette filled with 0.1 N AgNO₃ solution (Titrisol, Merck).

The titration data are transferred to a PC for improved evaluation. The evaluation process is semi-automatic and determines the best fit between the experimental data and a theoretical titration curve.

The blank determination is carried out using benzoic acid, which is mineralized in a freshly purified platinum boat.

Single-point calibration is performed using chlorobenzoic acid and bromobenzoic acid standard materials for elemental analysis.

Determination Using Capillary Ion Electrophoresis

Capillary ion electrophoresis provides an alternative method for the determination of chloride and bromide. It is particularly useful when the separation and individual determination of different ionic species are required.

Following mineralization or, where appropriate, direct dissolution or extraction of the sample, the resulting solution can be analyzed by CIE. The ionic species are separated according to their electrophoretic mobility and subsequently detected and quantified.

CIE can also be used for the determination of other inorganic anions and cations, making it a versatile complementary technique for the analysis of ionic species.

Working Range

The analysis enables the recovery and determination of all analyte elements present in the sample. A wide range of technical products, including PVC (57 wt-% Cl) and biogenic materials (<0.05 wt-% Cl), as well as common organic compounds, have been investigated.

The limit of determination was found to be approximately 0.05 wt-% (500 ppm) using a sample amount of 6-10 mg per analysis. The measurement uncertainty was estimated to be 0.02 wt-%.

Within the typical concentration range of organic materials, the uncertainty remains well below the 0.3 wt-% limit and has been demonstrated to be within the range of 0.1-0.2 wt-%.

Interferences

The handling of volatile substances or solvent mixtures can be problematic. Samples are normally manipulated using a capillary for weighing in order to minimize evaporation. When investigating mixtures, selective loss of the most volatile components may lead to errors. For the analysis of highly volatile liquids, the use of a Hamilton syringe has been successfully applied. This allows the sample to be introduced directly into the platinum boat immediately before entering the mineralization furnace.

It is well known that certain metals, such as Na, K, Mg, and Ca, do not completely react to form their corresponding oxides under the digestion conditions. Instead, they may retain a portion of the halide ions, which can lead to significant deviations of the analytical result from the true halogen content. The addition of vanadium pentoxide (V₂O₅) can minimize this effect and helps achieve results within the required accuracy of 0.3 wt-%. This occurs because the metals preferentially form vanadates, while the hydrogen halides are more readily released.

Mineralization of compounds and mixtures containing elements that form volatile halides at 900 ^oC, such as Al, Cu, Zn, Ti, and others, may also cause interferences. At high metal contents, reliable analytical results may no longer be obtainable.

Under such circumstances, hydrolysis of the substance or mixture under mild conditions may provide an alternative approach. The resulting digest can often be analyzed directly by argentometric titration or capillary ion electrophoresis.

Capillary Electrophoresis with Conductivity Detection. An Alternative to Ion Chromatography?

Capillary electrophoresis (CE) is a relatively new family of separation techniques that truly deserves the designation “high-performance separations” due to its remarkable separation efficiency. These methods are based on the principles of electrophoretic migration of charged species, as originally introduced in slab electrophoresis techniques.

Electrophoretic separation within a narrow-bore capillary allows experiments to be performed without the need for a gel or paper matrix. In slab electrophoresis, such a matrix is required to prevent convective processes from disturbing the migration of substances. However, the matrix inevitably causes varying degrees of interaction between the analyte species and its extensive surface area.

In an open capillary system, electroosmotic flow (EOF) becomes a useful mechanism for transporting analyte species through the capillary. In classical slab electrophoresis, however, this phenomenon is considered an unwanted interaction. This fundamental technique is known as capillary (free) zone electrophoresis (CZE).

All classical electrophoretic techniques have subsequently been adapted to capillary formats, resulting in improved control, efficiency, and speed. In addition, specialized method developments have led to powerful techniques such as micellar electrokinetic capillary chromatography (MEKC or MECC) and capillary electrochromatography (CEC), which now address application areas traditionally dominated by high-performance liquid chromatography (HPLC).

As in the early development of HPLC, detection remains a major area of technological advancement. UV-visible diode array detection is currently the standard approach, and several strategies have been developed to improve sensitivity, including bubble-cell and Z-cell designs. Almost every detector used in HPLC has also been investigated for CE applications, although not all have yet become commercially available.

In 1995, ATI-Unicam introduced a conductivity detector module for CE. Prior to this development, non-UV-absorbing ions could only be detected by indirect UV detection, which limited the choice of buffers according to detection requirements.

A CE instrument equipped with conductivity detection appears to cover application areas traditionally addressed by ion chromatography (IC). Therefore, the technique was introduced as capillary ion electrophoresis (CIE). The advantages of CIE compared with IC must be evaluated in detail with respect to sample type and concentration range. In general, CE-based approaches offer higher robustness and faster analysis times but may provide lower sensitivity compared with IC.

Conductivity detection is generally applicable for the identification of charged molecules and complexes, as conductivity is a fundamental property related to ionic mobility.

Historically, CIE was introduced in the Microanalytical Laboratory in 1995 using a CRYSTAL 310 Capillary Electrophoresis system equipped with a CRYSTAL 1.000 Conductivity Detector (ATI–Unicam, 1995). This instrument is no longer in use. The technique is currently performed using an Agilent 7100 Capillary Electrophoresis system.