The catalytic stratospheric ozone destruction by CFC's and other compounds is almost certainly too complex to fully replicate in a laboratory with current technology. The only laboratory of proper scale and conditions is the earth's atmosphere as a whole (no pun here, seriously ;).
Even the 1995 Nobel prize winners Crutzen, Rowland and Molina had made serious omissions from their original work in the 70's that are now known to be critical to the mechanisms of stratospheric ozone depletion. While Crutzen is pretty much a pure theoretician, Molina and Rowland did much of their work in the laboratory and combined that work with atmospheric measurements. The bottom line is that only pieces of the whole process can be controlled in a single or set of experiments. One important factor, for example, that was omitted from their original work was the role played by condensed-phase surface process in mediating catalysis by active chlorine. Specifically, without the different types of polar stratospheric clouds (i.e. water ice, nitric acid trihydrate, liquid supercooled solutions of sulfuric acid, nitric acid and water, and other complex matrices) catalytic ozone destruction does not occur. So, although the net of their predictions in the 70's was correct, it was the real-world observations decades later, not laboratory confirmations, that sent them to Stockholm for their prize.
In 2010, an article was published called "Key compound of ozone destruction detected; Scientists disprove doubts in ozone hole chemistry.", Karlsruhe Institute of Technology (KIT). ScienceDaily. ScienceDaily, 22 July 2010, excerpt available here. The summary states:
For the first time, scientists in Germany have successfully measured
in the ozone layer the chlorine compound ClOOCl, which plays an
important role in stratospheric ozone depletion. Doubts in the
established models of polar ozone chemistry expressed by American
researchers based on laboratory measurements are disproved by these
new atmospheric observations.
This represents an even more recent case in which rigorous laboratory attempts to simulate stratospheric ozone depletion processes failed in the face of experiments and measurements made in-situ in the "real world" laboratory.
The article further points out:
The ozone hole above the Antarctic and the destructive role of
chlorofluorocarbons (CFC) and their decomposition products have become
a synonym of both global environmental problems and their solution by
concerted agreements worldwide. Scientific fundamental research into
ozone chemistry of the atmosphere was the basis of international
agreements, such as the Montreal Protocol of 1987, which has put
limits on CFC production. The success of the political implementation
of these scientific findings is reflected by the fact that the
chlorine content of the atmosphere and, hence, the ozone destruction
potential recently started to decrease slowly.
So, the upshot here is that it was primarily in-situ experiments and measurements that led to predictions of catalytic chlorine mediated ozone destruction being later confirmed by measurements. The same was true for formulation of a solution to the problem, which again in-situ measurements confirmed.
In short, laboratory experiments have repeatedly been outperformed by in-situ experimentation and observation in the case of stratospheric ozone depletion studies. The same is true for many complex, global chemical and physical process; it simply is not always feasible to re-create a system of large scale and complexity in a laboratory.