[Postscript version]
This page was last updated on August 26 , 1998.
A primary goal of the September proton run is to study lambdas similar to those observed during the 1998 heavy-ion run. There is interest in developing a reasonably efficient means to trigger on lambdas that decay, inside the DDC, into a proton and pi-minus. The TOF array is well-suited to detect the daughters of these fiducial lambdas. Simulations were performed to determine the regions on the TOF array where the daughters of fiducial lambdas, and the noninteracting proton beam, would hit. Results and appropriate summaries are presented.
The slat widths (hole widths) are different for different regions of the array, though the hits-vs.-TOF hole histograms shown below do not reflect this:
| holes | slat width |
|---|---|
| 1-26 (beam-left wall) | 5 cm |
| 27-61 (central wall) | 1.7 cm |
| 62-100 ( " " ) | 1 cm |
| 101-186 ( " " ) | 1.7 cm |
| 187-200 (beam-right wall) | 5 cm |
A plot showing which slats of the TOF array would be hit by the noninteracting proton beam is shown below, for 15 GeV and 12 GeV beams, and for varying sweeper-field values.
The locations on the TOF array where the daughters of fiducial lambdas would strike is shown in the plot below.
For the pions from fiducial lambdas, the 26 slats that comprise the beam-left wall will collect 60 percent of them. In order to collect that many protons, 34 slats, in the range from hole 72 to hole 105, would be necessary. This would be one advantage in triggering on pions rather than protons.
Another advantage is that the pion distribution is far removed from the noninteracting beam. For a lower sweeper field, it becomes more difficult to distinguish beam protons from lambda protons.
In the next graph below, the momentum of the parent lambda is plotted against the TOF hole where its daughter proton or pion hit.
The bulk of the lambdas whose pions strike the beam-left slats are in the range from ~10 to ~20 GeV. When the lambdas with momentum above 20 GeV are cut away, the pion-hit efficiency dropped to 50 percent.
The same cut, applied to proton hits for holes 72 to 105, yielded a similar reduction in proton-hit efficiency.
In an attempt to improve the lambda acceptance for the proton run, the target may be moved closer to the DDC front face.
Before re-running the fiducial-lambda simulations, the angle of the 15 GeV beam at the (-8,0,253) target position was increased, such that the beam would just touch the beam-right front corner of the DDC. (The original beamline would pass about 7 cm to the beam-right of the DDC right front corner.) To produce that beam trajectory, the beam angle at (-8,0,253) was increased to 6.5 degrees.
The location of the noninteracting beam was determined for this larger beam angle. The results are shown below.
This new location for the proton beam is well within the within the distribution of protons from lambdas. The lambda-daughter distributions on the TOF array (accessible below) will show this.
With this redefinition of the beamline, fiducial-lambda event files were then produced for five points along this line: at the original (-8,0,253) target, 1/4 of the way to the DDC front face, 1/2-way, 3/4-way, and right at the DDC front face.
The fiducial-lambda acceptances increase in this fashion with target location:
| Target location | Pct. of lambdas decaying in DDC |
|---|---|
| (-8,0,253) | 1.6 % |
| (-1.875,0,284.375) | 2.8 % |
| (4.25,0,315.75) | 5.1 % |
| (10.375,0,347.125) | 10.1 % |
| (16.5,0,378.5) | 43.8 % |
This is one reason why the original beamline (with 3.16-degree beam angle) was not used: even when the target was 3/4 of the distance from the (-8,0,253) target to the DDC front face, the lambda acceptance was only 1.3 percent.
The distributions of the lambda daughters for these target locations were determined for these five target positions. The first row corresponds to a target at (-8,0,253), the second 1/4-way to the DDC, the third 1/2-way, the fourth 3/4-way, and the fifth row refers to the case where the target is at the DDC front face.
The postscript files of daughter distributions (for individual target positions) are available by way of the links below.
Shown below are the efficiencies for detecting lambda pions on the beam-left slats:
| Target location | Pct. of lambda-pions hitting BL slats |
|---|---|
| (-8,0,253) | 63.1 % |
| (-1.875,0,284.375) | 53.1 % |
| (4.25,0,315.75) | 41.6 % |
| (10.375,0,347.125) | 25.8 % |
| (16.5,0,378.5) | 10.7 % |
Note that aiming the beam closer to the DDC (from the (-8,0,253) target) improves the pion efficiency slightly.
Shown in the picture below are plots of the parent-lambda momentum versus the hole hit by the daughter particles, for the five different lambda-production target locations. Again, the first row corresponds to a target at (-8,0,253), the second 1/4-way to the DDC, the third 1/2-way, etc.
The links below provide plots of parent-lambda momentum vs. daughter-hit location, for individual target locations.
For targets closer to the DDC, the lambda momenta are lower; most tend to be around 3 to 5 GeV by the time the lambda-production target is halfway to the DDC. The pions of these lambdas tend to miss the beam-left slats altogether.
However, as shown by the table below, the percentage of lambdas in the momentum range between 10 and 20 GeV whose pions hit the BL slats is roughly independent of target location. As far as the efficiency of a pion-trigger for lambdas in this momentum range is concerned, it shouldn't matter where we place the target.
| Target location | Pct. of 10-20 GeV lambda pions hitting BL slats |
|---|---|
| (-8,0,253) | 77.8 % |
| (-1.875,0,284.375) | 74.5 % |
| (4.25,0,315.75) | 75.6 % |
| (10.375,0,347.125) | 72.7 % |
| (16.5,0,378.5) | 74.2 % |
Does moving the target closer to the DDC provide us with significantly more lambdas in the 10-20 GeV range? The table below, showing the probability of a produced lambda being fiducial AND having momentum between 10 and 20 GeV, is an attempt to answer this question.
| Target location | Prob. of a produced lambda being |
|---|---|
| fiducial and w/ mom'n between 10 and 20 GeV | |
| (-8,0,253) | 0.8 % |
| (-1.875,0,284.375) | 1.1 % |
| (4.25,0,315.75) | 1.4 % |
| (10.375,0,347.125) | 1.5 % |
| (16.5,0,378.5) | 2.2 % |
In principle, we should be able to roughly double the share of 10-20 GeV lambdas upon which the BL slats can be used to trigger.
If low-momentum lamdas are what we seek, perhaps a proton trigger, consisting of the ~40 slats to the right of the beam spot (which is around hole 100 to 110) might be useful, provided that the backgrounds in this region are not too high. This question has yet to be investigated.
The production of fiducial lambdas and the tracking of their decay products to the TOF array were simulated, as was the tracking of 15 and 12 GeV proton beams through the E896 detector and field geometry.
With the original beam trajectory and full fields, a noninteracting proton beam with momentum between 12 and 15 GeV should hit the TOF array somewhere between holes 120 and 135. The distribution of protons from fiducial lambdas on the TOF array are maximally between ~holes 70 and 110. This is close to the region of the noninteracting beam, and more so if the sweeper field is less than 60 kG. In this regard, a trigger on pions from fiducial lambdas would be preferable.
The beam-left wall, with its 26 slats, can detect 60 percent of the pions from fiducial lambdas. To collect that many protons from fiducial lambdas, 34 slats ranging from hole 72 to hole 105 would be required. A pion-trigger would thus be easier to implement, from a hardware point of view.
With the 15 GeV proton beam rotated at the (-8,0,253) target so that the beamline is closer to the DDC, the pion-trigger efficiency can be improved slightly. As the lambda-production target is moved further downstream, the pion-trigger efficiency does not change significantly for fiducial lambdas with momenta between 10 and 20 GeV. At the same time, however, more lambdas between 10 and 20 GeV are produced, so there should be more fiducial lambdas in that momentum range whose pi-minuses would be detected by the BL slats. Thus, by rotating the beamline closer to the the DDC, and moving the lambda-production target some practical distance closer to the DDC front face (half way or three-quarters of the way), the pion-trigger efficiency for lambdas in the momentum range of interest should be improved.
Among the open questions not addressed in the above simulation results are the p-Be interaction backgrounds, and how they would affect either a pion-trigger or proton-trigger. The momentum spread in the beam, and its effect on the distribution of daughter pions and protons, has not been investigated. Uncertainties in the sweeper and analyzer field values may affect the lambda-daughter and noninteracting-beam distributions as well (although a lower analyzer field value may actually improve the pion-trigger efficiency). Finally, the fiducial-lambda momentum range of interest would determine whether a TOF lambda trigger is used, and which slats would comprise it.
In the picture below, the transverse momentum of the parent fiducial lambda (in GeV) is plotted against the TOF hole where its daughter proton or pion hit, for the five different lambda-production target locations.
The transverse momentum of the lambda is defined here to be that component of its total momentum which is perpendicular to the beam trajectory at the point of production of the lambda. For each target location, the beam trajectory is different, so in each case, the transverse momentum is calculated with respect to a different axis.
In comparing the above picture with the total momentum-vs.-hole hit plot, it seems that the hit locations of high-pT lambdas coincide with hit locations of lambdas with high total momentum.
No calculations of acceptances for lambdas of a particular range have been made yet; the "raw" pT distributions are plotted here to give an idea as to what transverse momenta can be expected.
