Here’s a tip: If you’re going to claim you were first to discover something, even though you know you weren’t, don’t publish your claim in the same journal where the first finding appeared. Oh, and don’t ask the researchers who made the first discovery for help along the way.
Those, perhaps, are the cynical lessons from a retraction notice that appeared last week in the Journal of Chemical Physics:
The authors wish to retract the article1 because the CN(X2Σ+) + He potential energy surface utilized for the cold collision studies was misrepresented as having been newly developed in this laboratory. The potential energy surface for this system was originally calculated and published by Lique et al.2 We apologize for not properly referencing the previously published findings of Lique et al. or obtaining the necessary permissions to reproduce their potential energy surface in the text or caption to Fig. 1.
1. E. Feng, X. Shao, C. Yu, C. Sun, and W. Huang, J. Chem. Phys. 136, 054302 (2012).
2. F. Lique, A. Spielfiedel, N. Feautrier, I. F. Schneider, J. Klos, and M. H. Alexander, J. Chem. Phys. 132, 024303 (2010).
The paper has been cited just once, according to Thomson Scientific’s Web of Knowledge.
François Lique, first author of the original paper, tells Retraction Watch he became aware of the duplication because he reads the Journal of Chemical Physics and thought the paper had results that would be of interest to his research. When he read it, he realized it was the same work he’d done two years earlier. Even worse, he remembered that the authors had contacted him for some help — which he and his co-authors were happy to give.
If the publication had been in a journal that he didn’t read, “maybe I would not be aware of the duplication,” he wrote in an email.
So, the study was identical and the resuls too? Unlikely. It was not plagiarized, so I don’t really see the point. At worst, it is a replication of the original study. Something does not square.
It was not a duplication of the previous study – it was “a new discovery”!
The problem wasn’t a study as such, it was an analytic technique which the retracting author had claimed to be the first to develop. That’s different from conducting a study protocol on another batch of subjects…
I haven’t read the paper in great detail, but where exactly do they claim that they were the first to develop their method? Also, I find the retraction notice to be a bit weird. Why would they need permission to publish their Fig. 1? There is no copyright to the solution of an equation, nor to its graphical representation by a standard computer program.
I haven’t even looked at the paper. I was just reading the retraction notice literally. If they *didn’t* claim to be first, then you’re right, where’s the problem? As you say, there is no copyright to an equation…just bragging rights. Maybe Lique is a biggie. On the other hand, if the whole point of the paper is the equation(or computer program) then it’s duplicative regardless of copyright.
Yup, very strange. They may have lifted the data but rewritten the text and generated their own figures. As far as that may be possible, of course.
Or perhaps François Lique is an obnoxious biggie in the field and nobody wants to piss him off. I have no idea!
Perhaps, if you have no idea, you should do the few key strokes of work required to clarify the matter (as the next writer did). That is, do the required work to avoid creating apparently inaccurate, easily extracted statements such as the “François… field”. In this age of easy text mining. At this site especially. Prof PNP
The biggie in the list of authors is Millard Alexander. Dr. Lique was a post-doc in his group, Dr. Klos is a current member of the group. In carrying out such calculations the most important and first step is to develop a potential energy surface which describes the system. The Chinese group did not develop their own but used that found by the Alexander group. They also pretty much copied the word salad.
Here are a couple of paragraphs from each paper:
The CN–He system can be described by the usual Jacobi
coordinates: R, the distance between the He atom and the
center of mass of the CN molecule; r, the CN internuclear
distance and , the angle between R and r. The CN bond
defines the z axis and we assume that the CN–He system lies
in the xz plane. For low-energy rotational excitation, we determined
an accurate PES by means of partially spinrestricted
coupled cluster calculations with single, double,
and perturbative triple excitations RCCSD T .15,16 We kept
the CN bond distance r frozen at its experimental equilibrium
geometry re=2.214a0 Ref. 17 . We used the aug-ccpVQZ
hereafter denoted aVQZ basis set of Woon and
Dunning18 for the three atoms, augmented by the bond functions
optimized by Cybulski and Toczylowski19 placed at
mid-distance between the CN center of mass and He.
The wave function which asymptotically correlates to
CN X 2 + +He system is of 2A symmetry. The CN A2
state lies approximately 9250 cm−1 above the X state. In the
presence of the He atom, the degeneracy of this state is
lifted, giving rise to two adiabatic electronic states, of A and
A reflection symmetry. As was already discussed by Werner
et al.,20 the two 2A states, which correlate at large R to
CN X and CN A , mix strongly when the CN bond is
stretched slightly beyond its equilibrium distance.
The He–CN system can be described by the usual Jacobi
coordinates: R, the distance between the He atom and the
center of mass of the CN molecule; r, the CN bond distance
and θ, the angle between R and r. θ = 0◦ corresponds to the
He–N–C collinear configuration. The CN bond distance was
frozen at its experimental equilibrium geometry (2.214a0).37
The asymptotic wave function correlated to CN(X2 +) +He
system is of 2A symmetry. The CN (A2 ) excited electronic
state lies approximately 9250 cm−1 above the X state. In the
presence of the He atom, the degeneracy of A2 is lifted,
giving rise to two adiabatic electronic states of 2A and 2A
reflection symmetry. As was already discussed by Werner39
the two 2A states correlate to CN(X) and CN(A) at large R
when the CN bond is stretched slightly beyond its equilibrium
distance. To be sure to obtain the energy of the lowest 2A
state, we adopted the strategy suggested by Lique et al.38 The
complete active space self-consistent field (CASSCF) calculations
were first performed for the lowest three electronic
states. From the orbital saved in first step, we subsequently
carried out RCCSD(T) calculations for the X2A state. At all
calculated geometries we checked that the dominant CASSCF
configuration for the X2A state is greater than 0.9 and the
T1 diagnostic is smaller than the usual cutoff value of 0.02
(Ref. 40) to ensure that a single-reference description is valid.
It gets worse.