TAP experiments

TAP Hicroreactor and TPROX Studies. In an examination of the role of oxygen in the reaction, it is useful to decouple the redox reactions. In separate TAP experiments we pulsed a 50-50 blend of... 

Temporal analysis of products (TAP) reactor systems enable fast transient experiments in the millisecond time regime and include mass spectrometer sampling ability. In a typical TAP experiment, sharp pulses shorter than 2 milliseconds, e.g. a Dirac Pulse, are used to study reactions of a catalyst in its working state and elucidate information on surface reactions. The TAP set-up uses quadrupole mass spectrometers without a separation capillary to provide fast quantitative analysis of the effluent. TAP experiments are considered the link between high vacuum molecular beam investigations and atmospheric pressure packed bed kinetic studies. The TAP reactor was developed by John T. Gleaves and co-workers at Monsanto in the mid 1980 s. The first version had the entire system under vacuum conditions and a schematic is shown in Fig. 3. The first review of TAP reactors systems was published in 1988. 

In the same publication, a method for the parallelization of TAP experiments was also indicated. It was stated that ...high-throughput transient kinetics carried out in addition to high-throughput catalyst synthesis and testing both accelerate the search for new catalytic materials and bring fundamental insights into reaction mechanisms. ... 

This mechanism has been supported by the XPS investigations of Coulston et al. (109) and TAP experiments carried out by Rodemerck et al. (110,111) and Lorences et al. (20). These investigators observed that, in the absence of sites, no MA was observed, and the main reaction product was furan. When sites were present, MA was produced. However, the important sites were found to be short-lived isolated surface V " species rather than bulk VOPO4 phases. 

A novel approach for characterising carbon catalysts by TAP experiments... 

IR (5. ) and temporal analysis of products (TAP) (54) have been used to investigate the origin of the improved performance of fhe Pt/MO catalysts in CO oxidation. The TAP experiments shown in Fig. 12 demonstrate that the high activity of Pt/Co0 /Si02 in CO oxidation is related to the absence of CO inhibition effects at low temperatures. On the basis of these results it was proposed that CoO, is the suppher of O, which reacts with CO adsorbed on Pt. It is hkely that the reaction takes place at the Pt-CoO, interface. 

The results to be expected from a typical TAP experiment have been simulated (35) for a simple irreversible adsorption and are shown in Fig. 2. If there is no adsorption, the pulse at the reactor outlet is represented by curv e A. For > 0, some of the inlet pulse (N a moles of A) remains on the catalyst, and the amount is proportional to the difference between the areas under curve A and curve B, for example. Figure 3 shows what happens with reversible adsorption. For fast adsorption and slow desorption, there may be two peaks as shown by curve C. After a sufficient length of time, all the gas that was initially adsorbed will have left in the exit peak. Cleaves et al. (35) also show that for values of k. and kj that are sufficiently high so that the gas and adsorbed phases are everywhere in equilibrium, the response curve has the shape of curv e A but the peak height is reduced to 1.85/(1 + Keq). 

The difference in the behavior of Ni/Si02 and that of Ni/La203 has been shown by TAP experiments performed in the pump-probe mode at 600°C... 

Fig. 37. TAP experiments. Peaks of CO2 formed over polycrystalline Pt at 73°C caused by multiple pulses of O2 over an initial CO-covered surface (after 214).

The authors thank Dr. H.W. Zanthoff for his help in TAP experiments as well as helpful discussions. Financial support by the Bundesminister fiir Bildung, Wissenschaft und Technologie (grant-no. 423-4003-03D0001B0) is gratefully acknowledged. 

Table 1. Composition of gas mixtures used in TAP experiments (molecular amounts relative...

It was found that the amount of N2 (and also N2O) formation decreases with increasing offset time At from 0 to 1 second for the lean mixture B. The amoxmt and peak shape of the N2 formed during the NO pulse at At = 1 second are identical to those observed when NO is pulsed over a preoxidised surface. This indicates that one second after the propene/Oa pulse no residual carbonaceous species are present on the surface which can reduce NO. When the next propene/02 pulse enters the catalyst, two seconds after the NO pulse, still some N containing adspecies are on the surface as some N2 (and also N2O) formation is visible. At At = 0 and 0.01 seconds more oxygen leaves the catalyst after the propene/02 pulse then at larger offset times. Also more N2 is formed during the NO pulse at At = 0 and 0.01 seconds. This effect can be caused by a competition between O2 and NO for a direct reaction with propene or reaction products of propene. Another possibility is that O2 and NO compete for reduced adsorption sites. Rottlander et al. [14] recently reported similar results with TAP experiments on a Pt/ZSM-5 catalyst. They proposed that carbon containing surface species, formed from propene, are mainly responsible for the NOx reduction at T < 600 K. 

With the stoichiometric mixture A the amount of N2 and N2O formed was independent from the time interval between the propene/02 pulse and the NO pulse. Burch et al. [3] observed from similar TAP experiments with lean mixtures of propene/02/NO that the N2 yield did not change significantly with increasing At. 

TAP experiments are perfonned at operating conditions which are quite far from operating conditions in automotive converters pulse conditions, low pressure (lO" - 10 Pa) and low absolute concentrations. To find out in how far conclusions fi-om TAP experiments are still applicable to the automotive converter, correlation experiments were performed on the model gas test setup using gas compositions as given in Table 1. 

Figure 6 shows the NOx conversions and N2 selectivities for both model gas and TAP experiments with the stoichiometric and lean mixture as function of the temperature. The TAP measurements were carried out by simultaneously pulsing both valves. No NO2 formation was observed in TAP experiments in contrast with model gas experiments, therefore conversion and selectivity were calculated for the reaction of NOx (NOx = NO + NO2) to N2 and N2O. The equilibrium of NO and NO2 in O2 is given by ... 

When the temperature scale for the TAP measurements is adjusted to higher temperatures by approximately 100 K, a quite good correlation between the model gas and the TAP experiments is found for both the NOx conversion and the N2 selectivity and for both the stoichiometric and the lean gas mixture. The NOx conversion increases for the stoichiometric mixture with increasing temperature. The NOx conversion for the lean mixture shows a maximum in conversion at approximately 575 K (model gas temperature). The selectivity towards N2 shows for both gas mixtures a minimum around 575 K. This is of course equivalent to a maximum in N2O selectivity. In contrast to the conversion of NOx, the selectivity towards N2 is not strongly depending on the composition. From these figures it appears that the higher O2 concentration in the lean gas mixture compared to the... 

Further it is shown that the NOx conversions and the selectivities for N2/N2O observed within TAP experiments correlate well with the results from model gas tests. Therefor the mechanisms of N2 and N2O formation are probably identical in both model gas and TAP experiments. 

TAP experiments were also carried out during this start-up period. The initial CH4 pulses at 973 K showed mainly a formation of CO, H2O and small quantities of CO2. Pulses of CHt were continued until the formation of H2 and C2H4 was observed. Fig. 2A shows the height normalised responses of CH, H2, Ar, C2H4, CO and CO2 from a 9/1 CHi/Ar pulse. The CH4 conversion was smaller then 5% with approximate relative selectivities towards CO, CO2 and C2H4 of 90, 10 and... 

The stoichiometric reduction of both external and internal molybdenum oxide phases in the early stage of the reaction was clearly observed with the formation of CO/CO2 and H2O, accompanied with the formation of C2H4 and further on of CeH, either by NSSTK in flow mode or by TAP experiments in pulse mode. The modelling of the TAP experiments showed that CO and CO2 are formed simultaneously. 

Nikuradse s conclusion on the roughness effect in the laminar region was disputed by Kandlikar [3]. It was shown that the experimental uncertainty in Nikuradse s data was of the same order as the pressure drop measurements between the two adjacent pressure taps. Experiments conducted [4] clearly show a roughness effect on friction factor and transition Reynolds number. A comprehensive summary of the research in this field is given by Kandlikar et al. [4, 5]. 

---