Some recent scandals concerning the chemical composition of the exhaust gases of certain classes of cars seems to suggest the necessity to propose an openly accepted and directly verifiable method to precisely state the exact percentage of the nitrogen oxides in the exhaust gases of the vehicles. Here a possible procedure is proposed, in the ambit of a treatment that is not certainly meant to be comprehensive, particularly in relation to all the possible chemical species involved and of all the specifically applicable analytical variants, still it is intended to represent a conceptually coherent and rigorous method.
Conceiving reliable analyses for a vehicle motor fumes does certainly not represent a simple task. One has to consider several factors, like for instance firstly the quality of the fuel, that should obviously remain absolutely the same during the whole scanning process of all the various classes and types of motors working indeed with the same kind of it. Probably such an aspect represents by far the most important step in stating the quality of the exhaust gas fumes. That is why one should always, during these measurements, fill the tanks of the tested motors absolutely by stocking up with fuel from the same common reservoir.
Usually the cubic capacity of the motor, its power expressed in horsepower, the polluting category to which it belongs and the like are normally parameters that can be reasonably used to classify the vehicles exhaust gases to be tested according to common groups of cars.
Another crucial factor during the same analyses should be represented from the fuel consumption per time unit. Such a goal requires the inescapable need of being able to observe the progressive consumption on finely graduated transparent cylinder tanks, in order to precisely put directly in relation the arisen amount of pollutants with the one of the fuel consumed.
Indeed in any case the most environmental friendly hydrocarbon fuel based motor should always maximize the production of carbon dioxide, CO2, in relation to the correspondingly consumed amount of fuel during the motor functioning, being the 100% of the equivalent chemical conversion of the former into the latter the ideal level of the perfection to be reached. Among other things the carbon dioxide represents the organic product generated from a motor that can directly enter into the planetary process of the photosynthesis. In other obvious words it is perfectly compatible with the environment even if it comes from a combustion motor. Otherwise all the other hydrocarbon by-products, arising in turn actually from an incomplete combustion, real complex mixtures of compounds in form of volatile vapours – this one another highly complicating factor from an analytical point of view -, oils and carbonized powers an so on represent instead, in their form, poisonous and dangerous chemical matter for the nature and for the biological health.
During the analyses of the car exhaust gases the general conditions of detection should remain, as already pointed out, absolutely the same for all the checked vehicle-models of all the trademarks on the market belonging to the same polluting category and functioning with the same kind of fuel. About such an aspect one should indeed care almost in a maniacal way: it does represents completely the key prerequisite for a serious detection, which means systematically stocking up for the tanks of the vehicle to be examined with the fuel being dispenced from the same common reservoir. In fact, ideally, but also necessarily, such measurements should be carried out practically as much in a quick series as possible, performing repeated cross checks, avoiding absolutely to refill the main common fuel reservoir with additional fuel, even if it consists of the same quality or kind, - that is, for every refilling with new fuel of the main common fuel reservoir, one should retest correspondingly all the same motors again - and being sure to have employed absolutely unvaried and unaltered instruments, pipes, the electronic devices during all whole analysis process. The motor to be analyzed should work in absolutely standard regimen conditions, as for instance concerning the average motor performance. Whether the checked engine refers to a one connected to a car “on tires” or not represents another important factor, because a motor tested in the lab will behave quite differently, already in terms of consume, from a motor “on the street”, where other kinds of frictions, assuming in any case the vehicle completely unloaded, can contribute to influence the average consumes.
A fundamental principle in such analytic measurements should consist in keeping the chemical background as much as possible common and unvaried, independently from the percentage composition and the chemical pattern of the mixture to be analyzed. Otherwise unpredictable interferences might arise. Such a goal, strictly speaking, with the vehicles exhaust fumes is actually virtually impossible to be reached, because already the composition of the air and its whole amount too leaving the motor after the combustion process varies from a car model to another, either because some element previously present, like for instance the oxygen, is very differently consumed during the fuel combustion process, either because of the different carburetion conditions and engine types, even by remaining the same the overall final quantity of consumed fuel. Furthermore the time of the analysis represents also a crucial variable, that should remain absolutely the same for all the various specific detections, as the nitrogen oxides are in thermodynamic equilibrium with the respective elements (nitrogen, N2, and oxygen, O2), and they could gradually interconvert in them in the meanwhile. That is the reason why the duration of the analysis should strictly remain exactly unvaried during all the official tests for every correspondingly tested vehicle. During the various detection operations it is obviously assumed that the exhaust gases do not enter in contact with any kind of catalyst, which in turn might alter appreciably the decomposition speed of the various nitrogen oxides.
So, after these introductory statements it is now possible to consequently describe a method that would in principle accomplish a sure estimation of the levels of the nitrogen oxides in the exhaust gas fumes, without forcefully the need to separate or isolate them from the other components, neither to rely on electronic automatic detections systems, which, in spite of being formally very precise, they might provide deceiving results because of masking superimposing substances, particularly if the fumes are not appropriately pre-treated, like it is following described, in order obviously to purely minimize the effects of the possible interferences. Furthermore the electronic measurements of the various sensors seem to be generally applied on the “flows” of the fumes, aspect, this, that might imply a big disadvantage in the precision as it will be shown following.
So, of the two following described methods, the first being very concrete, whereas the second instead for the moment to be considered only with speculative theoretical value, should allow a reliable and confident determination of the nitrogen oxides levels in the exhaust gases fumes. But before illustrating them it is firstly at this point necessary to divide the pollutants of the exhaust gases in two categories, according to the possibility to separate them experimentally following the here proposed procedure (always obviously assuming the complete absence of catalytically acting substances).
Group 1: less or moderately volatile hydrocarbons, possibly also variably chemically functionalized, that condense partly or completely either at room temperature, or also at 60 °C, the temperature here chosen for simplicity as model, as it will be described following, plus oils and carbonized ashes. So that it should be possible to separate them from the main mixture (with the exception of the at such temperature very volatile low chain hydrocarbon vapours) before carrying out the analyses of the gaseous components, and all that through a cautious careful simple filtering operation.
Group 2: the gaseous components of the exhaust gas mixture at room temperature, or for simplicity at 60 °C, temperature at which one can be absolutely sure that there cannot in any case be present any nitrogen oxides form, even very partially, in a phase different that the gaseous one. With other words all the components of this group will be present under these conditions either in the gas phase (including firstly the nitrogen oxides) or as very volatile vapours, ascribable to very short chain organic fragments, for instance ethylenic, like theoretically (that must be verified) the ethylene or the ethylenoxide, or acetylenic and so on plus the water vapour, H2O, that during the combustion arises stoichiometrically, the carbon monoxide CO, and possibly also various other quantity of chemical “inorganic” species, depending on the motor type, the purity of the fuel, the carburetion regulation and more. And obviously majoritarily the carbon dioxide, CO2, that together with the water vapour arises stoichiometrically during the combustion process. Among the atmospheric gases, particularly the oxygen will be present in variable minor amount.
So here below the following procedure is outlined, taking into account also that, as the key components to be searched for during such detections are always the same ones, that is mainly the nitrogen oxides in the gas phase, N2O, NO, NO2, N2O4, - this last one, the so called dinitrogen tetroxide, is known, in spite of being in equilibrium with the nitrogen dioxide, NO2, even in the gas phase, to be present near 100°C virtually only as NO2 (in any case, the higher is the temperature, the higher is the corresponding decomposition to NO2) –, all that should afterwards easily become routine to confidentially identify and quantify their presence and percentage. Additionally possible inorganic various gases and vapours, as already pointed out, could in principle arise during the fuel combustion conditions, depending on the functioning motor regimen and on the purity of the fuel, and these must be verified experimentally from case to case
Premised also that even though a plenty of analyses with the IR spectroscopy will have been accomplished until today on the nitrogen oxides, the procedure here proposed should simplify dramatically the detection process and contribute in reaching an accurate precision. It is obvious that, if feasible, the second method, here, differently from the first one, proposed just theoretically, would represent without doubts the real level of the perfection. But also the method A should reach a rigorous systematic quantitative importance.
Detection Method A: The simplest straightforward method, simply consisting by collecting the gaseous mixture of the fumes, to which most of the volatile hydrocarbons and the water have been eliminated (it adsorbs strongly in the IR), and to measure the IR spectrum of the whole mixture. So that, by classify unequivocally the appropriate IR bands references and by having an internal standard in the target mixture, as it is obviously usual during quantitative IR analyses, it will be eventually possible to confidentially state the right percentage of the nitrogen oxides. In case unexpected superimpositions occurred, certainly the Raman spectroscopy can be used complementary to the IR, also for further confirmations in the results. But before starting in describing the main method it is worth to remind that as the intensity of the IR bands is very variable, depending both on the specific chemical compound, both on which IR band of it is considered, it becomes obvious that the corresponding intensity must be put in relation either with the selected IR bands intensity of the other nitrogen oxides either with the IR band of the internal standard. An absolutely necessary calibration operation without which one would arrive to very wrong results. The ratios between the nitrogen oxides percentages will give their mutual relative presence in the fumes mixture. Instead the ratios respect to the internal standard would reveal their absolute concentration in the gas mixture contained in the correspondingly known volume under the pressure of one atmosphere (1 atm), because the concentration of the internal standard is known. And all that through the proper evaluation of the IR bands (peak area integration, absorbance and so on). Such state of things, by combining the correlation also with the Raman spectroscopy, as well as by minimizing the presence of interfering gaseous and vapour substances during the analysis, should bring to absolutely reliable and above all quantifiable results. And this will be possible if the IR or Raman bands, belonging with certainty to the nitrogen oxides, that is to the N2O, the NO and the NO2 (but also the N2O4 must be considered), are free from superimposing interferences of any kind, so that they can be acceptably resolved and unambiguously correlated.
The experimental Procedure:
So here following a quite practical procedure is described in order to effectively carry out analytical measurements with the lowest possible extent of interference and with confident precision:
1) Collect the exhaust gases of the specific vehicle quantitatively. They should be then directed to the first piston-cylinder system, in which the external pressure acting on the gaseous mixture is absolutely equal to the atmospheric one (1 atm). In other words the piston should be free to flow along in the cylinder completely friction free, but also tightly in order to avoid any lost of the internal gas, so that at last the volume will turn out to be directly proportionally dependent on the whole internal amount of the gas. But particularly the measurable cylinder volume, associated with the by then known concentration of the internal standard, should allow the exact corresponding quantification of the nitrogen oxides. The piston-cylinder walls-system should also be able to conduct the heat efficiently, so that the initially warm exhaust fumes mixture will have the possibility to efficiently cool down.
2) Through a suitable quickly centrifugally moving blades system let now move fast the fumes inside the first piston-cylinder system in order to let them reach properly and stably a fixed temperature, for instance 60 °C, so that one can be sure, that at this temperature all the nitrogen oxides will be present in the gaseous phase and no other type of aggregation state can arise. Again it should be strengthened that the time factor represents an extremely important variable, because being the nitrogen oxides in thermodynamic equilibrium with the corresponding elements, their percentage, by different analysis times, might vary importantly. So what will be the right duration of the analysis ? Obviously this first phase is the time limiting one, due to the necessity of the cooling down of the gaseous fumes mixture until a fixed temperature, being the same in the ambit of all the analogous tests, the minimal common duration will be the one necessary to cool down the most slowly cooling down gas mixture of the fumes or the one that had the highest starting temperature, in order to guarantee analytical measurements carried out at the same temperature, with a unambiguously measurable volume and more in general under the same conditions.
3) Transfer very carefully all the content of the first piston-cylinder in a second one of the same kind (the same considerations done for the first piston-cylinder system should be valid for all the others ones) by filtering out the content, at the chosen fixed temperature, so that most of the at that temperature condensable vapours, plus the oils and the carbonized powders can be separated. By doing this the second piston-cylinder system will contain then only gaseous material plus a variable moiety of very volatile vapours attributable to very short chain hydrocarbon species arisen during the combustion process, and, very importantly, still the water vapour. Such operation will simplify already appreciably the gaseous mixture to be investigated for the presence of the nitrogen oxides, as it will then consist only in the chemicals of the above mentioned Group 2.
4) Transfer the content of the second piston-cylinder in a third one by appropriately dehydrating it through a very efficient dehydrating filter (but not having chemical influence on the nitrogen oxides levels) and at the same time filtering it, so that the third piston-cylinder system will just contain gases “water free”, and possibly very volatile hydrocarbon vapours, like ethylene, acetylene and other short chain hydrocarbons derivatives. Be sure that no water vapour is still there. The necessity to get rid off the water vapour is of crucial importance in such analyses, as the IR adsorption bands of the water vapour, that during the fuel combustion arises stoichiometrically, so in very big amount (like also the carbon dioxide, the CO2), seems to severely interfere with the IR spectrum of some nitrogen oxides, for instance in a region of the water IR scissoring bending band at about 1595 cm-1, usually of medium intensity. It is evident in this case that more accurate and complete numerical data are necessary.
[certainly the operations 3) and 4) could be also accomplished at once, but here, at least for simplicity, they are described separately]
At this point two alternatives would arise. The first quite straightforward and practical, the real operative one, which will be summarised immediately following, whereas the second one is for the moment intended as purely theoretical, but if it worked, it would represent the very ideal method.
So according to the first procedure, once that one has succeed to eliminate all the ashes, the oils, the less volatile hydrocarbon vapours and any trace of water, as well as in form of vapour, it will be possible to firstly directly measure very precisely the consumed amount of fuel (on finely graduated transparent tanks), in order to put it in correlation with the results obtained concerning the nitrogen oxides. At the same time the volume of the final piston-cylinder system, under the pressure of one atmosphere (1 atm) here for instance at 60 °C, represents at this point another very precisely measurable variable. At such a temperature all the interesting nitrogen oxides species to be analysed stay in any case completely in the gas phase and no risk of unpredictable condensation, what is more under the usual pressure of 1 atmosphere (where all the various boiling points of the interesting components are perfectly known), can take place, which should represent an essential requirement for the exact accuracy, as well as for the openness of the same analytical approach. So, at this point already two variable are known: the amount of the consumed fuel and the volume of the final piston-cylinder-system. Still such volume does not have automatically neither a rigorous meaning neither an absolute importance, because the air, among other things quite varied in its composition under such conditions (above all the oxygen, plus obviously the carbon dioxide), will be present in very variable amount, plus several unpredictable very volatile hydrocarbon vapours (see above) will be also there together with some inorganic gases or very volatile vapours species too. But, once that one has identified and calibrated the IR reference bands unequivocally resolved and correlated (again, remember that the intensity of the IR band is very variable according either to the chosen band, either to the type of compound, and must be very carefully put in relation, that is calibrated, with the actual concentration), plus an internal standard present with known concentration in the final piston-cylinder system of known volume, then two further results should become valid: the first being the relative ratio among the various nitrogen oxides, but also, very importantly, due to the presence of the internal standard at a known concentration, the absolute concentration of the various nitrogen oxides will become absolute. And all this under the highly strict condition of a completely unvaried time duration within the various detection test, as the nitrogen oxides stay in thermodynamic equilibrium with the respective elements, so that shorter or longer analysis times could drive to going astray results, so that such factor represents a crucial variable. Furthermore the total volume is also known (or precisely measurable), so that also the overall real amount of the nitrogen oxides will become known. It is evident that also the Raman spectroscopy can become very useful complementarily to the IR spectroscopy in this context, in case some unpredictable superimposition among the IR bands took place, but also as confirmation cross reference parameter about the ratio and the concentration of the nitrogen oxides. The preparation of external standards of the pure nitrogen oxides and the registration of the IR and Raman spectra at the same temperature is also highly advisable, together, that in any case prioritarily, with the exact calibration of the internal standard regarding the specific IR or Raman bands of the nitrogen oxides.
To conclude it is by far much more accurate to measure a certain final concentration of a pollutant present in a given measurable volume in which there are internal standards present at known concentration and also measurable pollutant amounts, that hardly relying on a “detected” presence of it in the fume flow. << If you want to state, as symbolic philosophical example, the volume of the wine contained in a glass, it is by far much more accurate simply to use a volumetrically finely properly graduated glass or container and to measure the amount of the in it poured wine than, blindly, without caring absolutely about the glass, “to estimate” the flow extent from the one coming out from the kept pouring bottle and to measure the time elapsed during the pouring operation ! >>.
And who knows that the here sensibly new proposed approach might actually revert some of the results obtained so far.
It is also very important to state whether any “cracking-like” process as by product of an anyway partial fuel combustion in the motor, causing the formation of short chain hydrocarbons (ethylenic, acetylenic an so on), in turn therefore very volatile and severely disturbing the detection operations, could be reasonably be actually minimised through the optimisation of some function performances of the motor, particularly the ones which improve the combustion extent process of the fuel during the motor functioning, like the carburetion, the air aspiration efficiency and its availability during the fuel combustion process, the spark lighting efficiency and the like.
Note that all the above operations should be previously run also using nitrogen oxides mixtures prepared especially for the purpose in known ratios (and kept at the temperature of 60 °C, also to record the IR spectrum at the same temperature of above), to carefully study all the various effects of the instrumental parts of the assembly proposed here, including the dehydrating material, on their chemical equilibrium and kinetic structural stability.
Some examples of interference:
A) For a chosen specific spectroscopic detection technique the right conceptual calibration represents an essential step in order to state the exact relative composition of a mixture or the concentration of selected chemical species. That is, if one chose for instance the IR spectroscopy for monitoring a gas mixture, one should consider that the intensity of the specific IR bands chosen as reference can strongly vary from compound to compound and that could lead to risky quantitative mistakes. So that, for instance, independently from the actual specific gas mixture of fumes under discussion, if one analysed a binary mixture of an alkyne plus an aliphatic ether, and tuned the IR detection, for monitoring the alkyne, on the medium-weak in intensity triple internal bond stretching band, to be found in most of the cases in the range between 2260-2190 cm-1 of the IR spectrum, otherwise, more in general, between about 2300 and 2100 cm -1, this one in turn usually several time less intense in general of the typically strong C-O asymmetric stretching band of an aliphatic ether, in the region between 1150 and 1070 cm -1 of the IR spectrum, then, actually, if one carried out the quantitative analysis of a binary mixture of an alkyne plus an aliphatic ether using such bands respectively as monitoring concentration reference for the alkyne and the above mentioned C-O aliphatic ether band as monitoring concentration reference for the ether, if such an aspect were not considered, one would certainly arrive to a very wrong result, because the alkyne, even if for instance present several times more than the ether in the mixture, might still possibly show either a smaller IR band integration area, either a lower absorbance than the one of the ether. In other words the minor component would appear to be the major.
B) Often some other chemical species possibly present in the mixture to be analysed might “simulate” the presence of another one and so in a “masking way” adding to them: if for instance one used the IR spectroscopy as analytical method for the detection of the various species, and so turned the attention on the dinitrogen oxide, N2O in the IR and on the corresponding calibration of the IR stretching band in the region of the IR spectrum between 1330 and 1250 cm-1, for detecting this gas, then one would in principle run the risk of blindly mistake a possible presence of acetylene, due to the IR combination overtones bands of the acetylene in the IR region between about 1380 and 1280 cm-1, but also of the methane, if present, due to the very strong sharp IR bending band at about 1300-1310 cm-1, just in the above reported region of the IR band of the N2O, so that a presence either of the methane and/or of the acetylene would subtly and deceivingly mask and add to the amount of the N2O, that would appear in this way to be present at much higher concentrations, even though its actual concentration in the mixture were much lower. Whether such chemical by-product species, taken here just as pure example for error-risk during such analyses, could arise as direct fragmentation products of a partial fuel combustion, or as consequence of not sufficiently pure fuel, or even as result of an immediately subsequent transformation, in presence of the abundant water vapour, in turn stoichiometric product of the combustion reaction, of possibly meanwhile originated equivalent “covalent chain like carbides”, to form methane and above all acetylene, all this must be verified experimentally.
Detection Method B: This procedure is here not meant as a real operative alternative to the detection method A, because it still must be tested. So, rather for the moment as a merely theoretical possibility, which nevertheless would represent for certain the very ideal case. It would consist in the so called “cryogenic gas solid or liquid chromatography” of the gaseous components present in the mixture. That means, a gas chromatographic process starting at extremely low temperatures at which most of the target gases will be present almost completely in the liquid or solid aggregation state, and so, provided that the hardware is suitable for such a challenge, then it should allow the selective separation of the searched specific gaseous components. If it worked such approach would represent the absolute level of the perfection that can be reached in such kind of analyses, able to selectively separate the various nitrogen oxides, and in this way to quantify them very precisely, as they could be evidently analysed separately, after their possible separation during the gas chromatographic process without the co-presence of the other interfering gaseous or vapour species. By such gas-chromatography the Helium, He, would represent probably the ideal carrier gas, as a consequence of it chemical inertness and of its physical constants (like fusion and boiling point). So that, after having injected the mixture at 60 °C (a subsequent higher heating of the injector would cause an dramatic alteration of the nitrogen oxides levels and what is more a quite risky overpressure) the starting temperature of the gas-chromatographic column would be settled at very low temperatures, for instance – 210/-196 °C (temperature range of the liquid nitrogen), but if practically feasible for such an instrumentation even lower, for instance starting temperatures at about – 250 °C, but the very ideal temperature could be identified probably after several attempts and corresponding observations, possibly on already known gaseous mixtures of the nitrogen oxides. All that during a carefully managed temperature programmed gas-chromatographic process. So that, as long as an overloading of the gas-chromatographic column did not arise, then such method should allow the separation of the main gaseous components of the nitrogen oxides present in the exhaust gas fumes. The extremely low starting temperature according to the already introduced principles, in the ambit of a temperature programming analysis, should prevent the gas-chromatographic gaseous components from migrating at the same speed of the carrier gas, which would make them practically undetectable. Again the time factor would turn out to be of crucial importance, so that no analysis should last more than another if the nitrogen oxides must be detected (due to the fact that they are in equilibrium with the corresponding elements).
During the gas-chromatographic process all the various components, including the very volatile hydrocarbons, could be in this way detected and identified, for instance through IR spectroscopy or Raman spectroscopy. In such a way the absolute perfect detection of the nitrogen oxides would be easily and directly reached.
A careful analysis of the boiling points of the various atmospheric gases should allow the confident separation of the nitrogen oxides. The noble gases Argon, Ar, Kripton, Kr, Xenon, Xe, and Radon, Rn, among other present in very low amount in the atmospheric air composition, would seem to likely create the most gas-chromatographic superimposition problems because of their boiling points, but due to the fact that they are not involved in the combustion process, their presence in the piston-cylinder system under one atmosphere (1 atm) of pressure would remain unvaried and constant, so not creating appreciable problems. Only the carbon dioxide, CO2, for analogous chemical-physical reasons and because of its copious (stoichiometric) production during the fuel combustion process is likely to cause the most difficult to overcome gas chromatographic superimposition hinder in this sense. But under this point of view progresses can be certainly done.
Nevertheless such kind of very low temperature, cryogenic gas-chromatography, would prevent the decomposition of the most sensitive gaseous components and potentially guarantee the very exact quantitative analysis and detection, for instance through IR or Raman spectroscopy, even if certainly the resolution of the eluting fractions of the analytes will be not of the same quality (broader peaks) of the usual gas chromatographic runs. In any case the feasibility of this second alternative must be taken with very big discerning and caution and needs to be verified directly. But it still does represent a challenging technological perspective for the future that is definitely worth the searches and the efforts to optimizing it.
* The essential parts of this article had been originally added, as enclosed text, to the article concerning the Global Warming, “The Global Warming: Causes, Remedies and Risks” on the following link: http://www.economistan.org/article/the-global-warming-causes-remedies-and-risks/ since the 07 October 2015, and published effectively on the following link: http://www.economistan.org/article/analysis-of-vehicles-exhaust-gases/ since the 3 Mai 2016, even though in the specific case only partially, because of technical reasons on the same web-site. The article appears here without bibliography.
** A preliminary literature search seems, in spite of a quite wide research extent in the ambit of the catalyst and the sensors for the detection of the nitrogen oxides, to exclude such kind of applications to detect the levels of the nitrogen oxides.