RESPONSE
Daniel Murnick has submitted a comment in defense of his earlier publications regarding the detection of radiocarbon with intracavity optogalvanic spectroscopy (ICOGS), in which he points out perceived shortcomings in our research (Carson et al. Reference Carson, Stute, Ji, Polle, Reboul and Lackner2016). Whereas we would expect insightful criticism from such an eminent scholar in the field, we were instead dismayed to read weak and unsubstantiated counterclaims to our findings.
His first criticism is that we should have demonstrated unequivocally that his earlier results could not be reproduced under identical conditions. We felt that this had already been accomplished in two different laboratories, at Groningen University and at Uppsala University (Persson et al. Reference Persson, Eilers, Ryderfors, Mukhtar, Possnert and Salehpour2013; Paul and Meijer Reference Paul and Meijer2015; Persson and Salehpour Reference Persson and Salehpour2015), but he arbitrarily disregards their findings. For what reason should we also disregard the invalidating results from other laboratories when no evidence is produced to discredit them?
Second, he claims that the theoretical basis for our findings was invalid. Our theoretical basis is an extension of the model he espoused in his own peer-reviewed publications. Compare what we present against his own model (Murnick et al. Reference Murnick, Dogru and Ilkmen2008). We expanded one of his products into an integral and placed rough bounds on his own otherwise unspecified term K, which “is a corresponding optogalvanic proportionality constant that depends on the details of the electric discharge” (Murnick et al. Reference Murnick, Dogru and Ilkmen2008). If our model is invalid because of the reasonable caveats of Bachor et al. (Reference Bachor, Manson and Sandeman1982), should this not also challenge his own published model?
Third, he claims that the crucial flaw in our experiment was that we used sawtooth modulation, “[making] the intrinsically narrow band CO2 laser effectively broadband.” Rather than provide evidence drawn from a comparable CO2 laser operated in a similar fashion, the support is claimed from an entirely different system (Genoud et al. Reference Genoud, Vainio, Phillips, Dean and Merimaa2015). The “lower sensitivity” found by Genoud et al. compared to Galli et al. (Reference Galli, Bartalini, Cancio, De Natale, Mazzotti, Giusfredi, Fedi and Mando2013) was attributed by Munick to the line scanning, not to any of the other potential sources of discrepancy. Genoud et al. was working at ~275 K at 20 mbar with a 3-mW quantum cascade laser, whereas Galli (2013) was operating at 195 K at 11.6 mbar with a 30-mW Ti:sapphire laser. Is it fair to disqualify our experimental system because two other systems, based on completely different principles with completely different equipment, had a differing results?
Fourth, he makes the unsubstantiated counterclaim, “In order to measure 14CO2 content with wavelength modulation it is best to use square wave modulation with a voltage equal to about ….” This statement is nothing but speculation. There is no precedent or experiment upon which to base this claim, thereby making it impossible to evaluate. Likewise with his insistence on the use of an optogalvanic reference cell instead of a photodiode: perhaps it would be better, but no evidence is presented to support this. If, for the sake of argument, we accepted the foregoing unsubstantiated counterclaims, which of the claims of our own report would still stand? We claimed that the optogalvanic measurements at the various 14CO2 lines were primarily the result of nearby 12CO2 lines resonances such as P(19)e and Q(13)e. We claimed that these lines constituted an unrecognized confounder in the previously published research on this topic. We claimed that this confounder was so large that it would have rendered detection of 14CO2 at the part-per-trillion level with the precision claimed was impossible. Given that none of these claims were challenged, we must assume that Murnick has conceded them. This again challenges his own claims (Murnick et al. Reference Murnick, Dogru and Ilkmen2008), which make no mention of these confounding factors.
Three research groups outside of Murnick’s laboratory have now attempted to reproduce his results and all three attempts have been foundered in their considerable efforts. With the exception of a few of Murnick’s own conference presentations, nothing has been presented beyond what was in Ilkmen’s thesis (Reference Ilkmen2009) that supports the initial promise of ICOGS. Murnick’s citation to his recent conference presentation (Murnick Reference Murnick2015) is unpublished; it is simply a review of the subject and appears to offer nothing new. Instead, a raft of peer-reviewed research has been produced documenting failure by other groups to recreate any sign of the resolving power promised in the prior publications and supporting the null hypothesis for ICOGS.
For the sake of brevity, we have focused only on the salient points of Murnick’s criticism. Regarding those parts of his comment that we have not discussed; differences of opinion are a natural part of the scientific endeavor and we encourage the reader to reach to their own conclusions on the basis of what has been written. We are thankful for his praise regarding some aspects of our experimental design. Perhaps his recommendations would be more helpful in other, more vital areas of scientific research.
