High-lying long-lived doubly excited states near ionisation thresholds of helium.

E Sokell*, A A Wills*, P Hammond*, M K Odling-Smee* and M A MacDonald@.

*Physics Department, The University of Manchester, Oxford Road, Manchester M13 9PL UK.

@Daresbury Laboratory, Daresbury, Warrington, WA4 4AD, UK



 

 
 
 
 
 

Background information.

In a conventional photoion spectrum one would normally expect a step increase in the yield of ions (or electrons) as an ionization threshold is crossed. In this experiment an unexpected peak was observed just below the N=1 threshold in Helium see figure 1. Two different experimental arrangements have been used to investigate this feature (figure 2 and figure 3).

All the data presented here was collected at Daresbury Laboratory using the Synchrotron Radiation Source (SRS).

[Image - Figure 1]
Figure 1 - Photoion spectrum of the Helium N=1 threshold



 

 
 
 
 
 

[Image - Figure 2]
[Image - Figure 3]



 

 
 
 
 
 

Results

Data collected with apparatus A is shown in figure 4. Unexpected peaks are observed below the He+(N=1) threshold in both the secondary electron and secondary ion channels. Peaks associated with the He(1s,np) series can be observed in both spectra. The relatively small intensities of the He(1s,np) series in the secondary ion channel combined with crude timing measurements suggest that photons impinging on the collision plate are less likely to produce secondary ions. For this reason the secondary electron channel is likely to be a mixture of photon and neutral excited atom signal, whereas the secondary ion channel is thought to be primarily due to neutral excited atom signal.

[Image - Figure 4]

A diagram of the timing windows can be seen below along with an example spectrum to demonstrate the TOF measurements.
[Image - Figure 7]
Figure 7 -Example TOF spectrum taken off resonance at h=24eV.

[Image - Figure 5]

Figure 5 shows data collected with the TOF apparatus (B). Photoemission from the He(1s,np) states produces a spectrum which clearly shows the series members up to n=8. Unlike the spectra in figure 4 there is no unexpected peak below threshold. In the background channel however a relatively large feature is observed.

[Image - Figure 6]

Unexpected features have also been observed below the N=2,3 and 4 thresholds in Helium. Figure 6 shows an example spectrum taken over the N=2 threshold.



 

 
 
 
 
 

Discussion

The simplest explanation of the unexpected peaks is that the natural lifetimes of the states are long. In the case of apparatus A, the neutral excited atom would have to survive for 70s in order to decay in the region beyond the charge particle suppression system. In the TOF apparatus preliminary analysis suggests that the particles responsible for the increase in yield in the background channel below the N=1 threshold have a lifetime of 250 ns. the peak is centred at an energy which corresponds to the n=19 member of the (1s,np) series which has a lifetime of 407 ns.

Initially excited states may fluoresce to the (1s,2s) metastable state which is long lived (1.9 ms). Below the N=1 threshold the members of the (1s,np) Rydberg series decay to this state with a branching ratio of 3%. Above the N=1 threshold doubly excited states normally decay by autionization. To decay via photoemission the doubly excited state has to have a lifetime against autoionization comparable with fluorescence lifetimes.

Collisional processes may also explain the observed features. Doubly excited states may be collisionally de-excited leaving them or the colliding atom in a metastable state. This process cannot occur below the N=1 threshold because there is no mechanism for carrying away excess energy. Other collisional exchanges can occur where small amounts of energy are exchanged enough to change l or n for high Rydbergs. This mixing increases the average lifetime of the collision product.

The presence of an electric field in the interaction region may be enough to induce Stark mixing which increases the lifetimes of the atoms by mixing levels of different l and, to a lesser extent, n.

In reality the true reason for the enhancement of these excited states probably lies in a combination of the above factors. These features have been observed in different experimental setups and are likely to be present in many other experiments. Above the N=1 threshold, doubly excited states which have highly correlated electrons, and hence long lifetimes, may play a considerable part in the physics of these features. For these reasons the features are of great interest for further study.