1: % ****** Start of file apssamp.tex ******
2: %
3: % This file is part of the APS files in the REVTeX 4 distribution.
4: % Version 4.0 of REVTeX, August 2001
5: %
6: % Copyright (c) 2001 The American Physical Society.
7: %
8: % See the REVTeX 4 README file for restrictions and more information.
9: %
10: % TeX'ing this file requires that you have AMS-LaTeX 2.0 installed
11: % as well as the rest of the prerequisites for REVTeX 4.0
12: %
13: % See the REVTeX 4 README file
14: % It also requires running BibTeX. The commands are as follows:
15: %
16: % 1) latex apssamp.tex
17: % 2) bibtex apssamp
18: % 3) latex apssamp.tex
19: % 4) latex apssamp.tex
20: %
21: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
22: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
23: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
24:
25: % Some other (several out of many) possibilities
26: %\documentclass[preprint,aps]{revtex4}
27: %\documentclass[preprint,aps,draft]{revtex4}
28: %\documentclass[prb,preprint,amsmath,amssymb]{revtex4}% Physical Review B
29: \documentclass[prb,amsmath,amssymb]{revtex4}% Physical Review B
30:
31: \usepackage{graphicx}% Include figure files
32: \usepackage{dcolumn}% Align table columns on decimal point
33: \usepackage{bm}% bold math
34:
35: %\nofiles
36:
37: \begin{document}
38:
39: %\preprint{SMU/01-05}
40:
41: \title{A Compact Apparatus for Muon Lifetime Measurement and\\ Time
42: Dilation Demonstration in the Undergraduate Laboratory}
43:
44:
45: \author{Thomas Coan}
46: \email{coan@mail.physics.smu.edu}
47: \author{Tiankuan Liu}
48: \email{liu@mail.physics.smu.edu}
49: \author{Jingbo Ye}%
50: \email{yejb@mail.physics.smu.edu}
51: \affiliation{Physics Department, Southern Methodist University, Dallas, TX 75275 USA}%
52:
53: %\date{\today}% It is always \today, today,
54: % but any date may be explicitly specified
55:
56: \begin{abstract}
57: We describe a compact apparatus that automatically measures the charge
58: averaged lifetime of atmospheric muons in plastic scintillator using
59: low-cost, low-power electronics and that measures the stopping rate of
60: atmospheric muons as a function of altitude to demonstrate
61: relativistic time dilation. The apparatus is designed for the
62: advanced undergraduate physics laboratory and is suitable for field
63: measurements.
64: \end{abstract}
65:
66: \pacs{Valid PACS appear here}% PACS, the Physics and Astronomy
67: % Classification Scheme.
68: %\keywords{Suggested keywords}%Use showkeys class option if keyword
69: %display desired
70: \maketitle
71:
72: \section{\label{sec:intro}Introduction}
73:
74: Measurement of the mean lifetime of muons produced in
75: Earth's atmosphere from collisions between cosmic rays and air nuclei
76: is a common experiment\cite{hist1,hist2,hist3} in advanced
77: undergraduate physics laboratories. Typically, a single scintillating
78: medium, massive enough to stop some useful fraction of the muons
79: impinging on it, is viewed by one or two photomultiplier tubes (PMTs)
80: that detect the pair of scintillation light flashes associated with
81: the entry and subsequent decay of a stopped muon. Histogramming the
82: time interval between the two flashes and then fitting the time
83: distribution with an exponential function yield the mean muon
84: lifetime. Various PMT readout and histogramming techniques have been
85: implemented to produce the decay time histogram. However, such
86: techniques tend to rely on relatively expensive and bulky electronic
87: instrumentation (e.g., NIM and CAMAC-standard modules) and stand-alone
88: multi-channel analyzers to generate the decay time histogram. We have
89: developed fully equivalent readout instrumentation, based on a complex
90: programmable logic device (CPLD\cite{cpld}), that is compact ($20\times
91: 25\times5\,{\rm cm^3}$), low cost and low power ($<25\,$W). The
92: readout instrumentation is easily interfaced to a laptop computer
93: to display and fit the decay time histogram.
94:
95: \section{\label{sec:det} Detector and Readout Electronics}
96:
97: We use a standard detector configuration, with no attempt made to
98: select only vertically traveling muons. A plastic scintillator in the
99: shape of a right circular cylinder ($15.2\,{\rm cm}$ diameter and
100: $12.7\,{\rm cm}$ tall) is viewed by a single 10-stage, $51\,$mm
101: diameter, bi-alkali photocathode PMT biased to nominal gain $3\times
102: 10^5$ attached to one end. Both scintillator and PMT are wrapped
103: carefully with aluminum foil and electrical tape to prevent light
104: leaks. The PMT is biased using a compact, commercially available DC-DC
105: converter\cite{emco} with negative high voltage (HV) applied to its
106: photocathode. To mimic events where a muon enters, stops and then
107: decays inside the scintillator, a blue light emitting diode (LED) is
108: permanently inserted into a small hole drilled in one end of the
109: scintillator. The LED can be driven by a transistorized pulser circuit
110: that produces pairs of pulses at a nominal repetition rate of 100 Hz
111: with an adjustable interpulse separation in a given pair from
112: $300\,{\rm ns}$ to $30\,\mu{\rm sec}$. For robustness and portability,
113: and so that no HV electrodes are exposed to students, the
114: scintillator, PMT, HV circuit and pulser are all enclosed inside a
115: black anodized aluminum tube $36\,$cm tall and $15.2\,\text{cm}$ inner
116: diameter. The cylinder is capped at both ends. Power to the HV supply
117: and pulser circuitry is provided by a single multi-connector cable and
118: the PMT signal is sent to the readout electronics module by a coaxial
119: cable. Potentiometers installed in the cylinder cap allow student
120: adjustment of the PMT HV and LED interpulse time separation. Provision
121: is made to monitor the HV with a conventional voltmeter and output of
122: the pulser circuitry is accessible by a coaxial connector on the cap.
123: The mass of the overall detector is $5\,$kg.
124:
125: The electronic circuitry to process PMT signals, perform timing,
126: communicate with the computer, and provide power to the detector is
127: mounted on one printed circuit board (PCB) located inside a single
128: enclosure of volume $20\times 25\times5\,{\rm cm^3}$. A block diagram
129: of this circuitry is shown in Fig.~\ref{fig:box}. Signals from the
130: PMT anode are coupled by the coaxial cable to the input of a two-stage
131: amplifier constructed from a fast current feedback
132: amplifier\cite{amp}. A typical raw PMT signal amplitude for muon decay
133: events is $100\, $mV into $50\,\Omega$ impedance. The amplifier output
134: feeds a discriminator with adjustable threshold and TTL logic
135: output. Students can monitor the amplifier and discriminator outputs
136: via BNC connectors mounted on the front of the electronics enclosure.
137:
138: \begin{figure}[ht]
139: %\rule{5cm}{0.2mm}\hfill\rule{5cm}{0.2mm}
140: %\vskip 2.5cm
141: %\rule{5cm}{0.2mm}\hfill\rule{5cm}{0.2mm}
142: %\hspace{3.5cm}
143: \includegraphics[scale=.50]{box.eps}
144: \caption{Block diagram of the readout electronics showing the
145: two-stage amplifier, discriminator, CPLD and I/O communications
146: circuitry. Signals are processed from left to right.
147: \label{fig:box}}
148: \end{figure}
149:
150: The discriminator output signal is processed by a CPLD, mounted on a
151: single PCB. A CPLD is a single, flexible integrated circuit (IC)
152: comprised of thousands of usable logic gates that can implement a wide
153: variety of digital logic functions. The chip includes programmable
154: interconnections and multiple input/output (I/O) pins, and can be
155: clocked at rates up to $\sim 100\,$MHz. Its behavior can be
156: extensively and reliably simulated before its controlling program is
157: downloaded into its electrically erasable programmable read only
158: memeory (EEPROM). Such ICs are reprogrammable and relatively cheap,
159: typically costing a few tens of dollars, making the inexpensive
160: implementation of a wide variety of digital logic signal processing
161: circuitry practical.
162:
163: The CPLD is programmed in a digital hardware description language
164: (VHDL) to function primarily as a timer. A logical ``yes'' output from
165: the discriminator, corresponding to an amplified PMT signal above
166: threshold, causes either the CPLD to start recording clock cycles from
167: the 50 MHz crystal oscillator that clocks it, or, if the CPLD had
168: already started counting clock cycles, to stop counting. In this way,
169: any amplified PMT signal above threshold can serve as a stop or start
170: timing signal. The CPLD is also programmed to reset itself if a second
171: ``yes'' does not occur within $20\,\mu$sec of the first one. This
172: simple logic scheme corresponds to our desired event scenario where
173: we have a flash of scintillator light corresponding to a muon entering
174: and stopping in the scintillator, and a second flash occurring when
175: the stopped muon decays.
176:
177:
178:
179: The CPLD formats data in a simple fashion. For successive PMT signals
180: above threshold and within the $20\,\mu$s timing window, the time
181: difference, in integral units of $50\,$MHz clock cycles, between these
182: two signals is recorded. Data of this type is ultimately histogrammed
183: and fit to an exponential curve by the laptop software to extract the
184: muon lifetime. For cases when there are no signal pairs within the
185: timing window, the CPLD merely records the integer 1000, corresponding
186: to the number of clock cycles in our timing window. All data are
187: subsequently sent to the laptop through either a serial or USB port.
188:
189: The CPLD I/O circuitry (see Fig~\ref{fig:box}) has two physical ports
190: to simplify interfacing to laptop computers. One port is a standard
191: serial port that follows the RS-232 protocol and that relies on a
192: dedicated RS-232 transceiver chip to shift RS-232 standard voltage
193: levels to low voltage TTL levels to communicate with the RS-232
194: interface module resident in the CPLD. The data transmission rate
195: between CPLD and laptop is 115~kbits/s. The other port adheres to the USB
196: 1.1 protocol and relies on a USB-FIFO (``first in, first out'')
197: translator chip to communicate with the FIFO interface module within
198: the CPLD. Data transmission rates in this case are 2.4~Mbits/s.
199:
200: Overall power consumption of the electronics module is less than
201: $25\,$W, sufficiently low that it can be powered in the field from an
202: automobile cigarette lighter.
203:
204: \section{Data Display Software}
205:
206: The laptop-resident software that displays and curve fits the decay
207: time histogram is written in the Tcl/Tk scripting language, an open
208: source language that permits easy implementation of graphical user
209: interfaces and that is compatible with Unix, Microsoft Windows and
210: Apple Macintosh operating systems. The laptop continuously examines
211: its own I/O port buffers for the presence of any data that the CPLD
212: has sent it. Any data consistent with PMT pulse pairs within the
213: timing window has the corresponding pulse separation time entered into
214: a decay time histogram. Data not corresponding to pulse pairs is used
215: to update various rate meters monitoring the frequency of PMT signals
216: above threshold. All data is then slightly reformatted to include the
217: absolute time in seconds when it was examined by the laptop before
218: being written to disk in ASCII format for easy human interpretation
219: and exporting to student written data analysis software routines. The
220: histogram and rate meters are displayed in real time for student
221: observation.
222:
223: The laptop software has provision for simulating muon decay by randomly
224: generating times according to an exponential distribution with a user
225: selectable lifetime. This permits students to practice their curve
226: fitting and lifetime extracting software routines on large simulated
227: data sets.
228:
229:
230: \section{Mean Muon Lifetime}
231:
232: A decay time histogram for muons stopping in our detector formed by
233: histogramming the time between two successive scintillator flashes
234: within our $20\,\mu$s timing window is shown in Fig.~\ref{fig:histo}.
235: Dots with crosses are data and the line is a fit to the data. This
236: histogram contains $28,963$ events collected over 480 hours of running
237: and contains $\mu^+\text{and}\ \mu^-$ decays as well as
238: background. The data is fit to the functional form
239: $\dot{N}(t)=P_1P_2\exp{(-P_2t)} + P_3$, characteristic of radioactive
240: decay with background. Here $\dot{N}(t)$ represents the observed
241: number of decays per unit time at time $t$ and the quantities $P_1,
242: P_2^{-1} \text{and\ } P_3$ are constants extracted from the fit and
243: represent an overall normalization constant, the muon lifetime and the
244: background level, respectively. These values are shown in the box in
245: the upper right-hand corner of Fig.~\ref{fig:histo}. The quality of
246: the fit is indicated by the low $\chi^2$ per degree-of-freedom
247: ($\chi^2/\text{d.o.f.}= 73/55$). The fit was done with PAW, a freely
248: available\cite{paw} fitting and plotting software package. Other
249: fitting packages return similar results. Due to the properties of an
250: exponential function, it is irrelevant that the muons whose decays we
251: observe are undetected when born outside the detector.
252:
253:
254: \begin{figure}[h]
255: %\rule{5cm}{0.2mm}\hfill\rule{5cm}{0.2mm}
256: %\vskip 2.5cm
257: %\rule{5cm}{0.2mm}\hfill\rule{5cm}{0.2mm}
258:
259: \includegraphics[scale=0.5]{mu_life.eps}
260: \caption{Decay time histogram for 28,963 events collected over 480
261: hrs. The dots with error bars are data and the line is a fit to the
262: data with the three-parameter function $f(t)=
263: P_1P_2\exp(-P_2t)+P_3$. The values for the fit parameters plus the
264: $\chi^2$ per degree-of-freedom are shown in the upper right hand
265: corner.
266: \label{fig:histo}}
267: \end{figure}
268:
269:
270: The extracted value of the mean muon lifetime $\tau= P_2^{-1}
271: =2.12\pm0.02\,\mu$s (statistical error only) is less than the free
272: space value $\tau_{\mu}= 2.19703\pm 0.00004\,\mu$s due to the
273: non-negligible probability that a $\mu^-$, but not a $\mu^+$, will be
274: captured into the K-shell of a scintillator carbon atom and then be
275: absorbed by its nucleus\cite{wardetal}. (The probability that a
276: stopped $\mu^-$ will be absorbed by a target atom of atomic number $Z$
277: is proportional to $Z^4$.)
278:
279: The extracted background rate of fake muon decays is inferred from the
280: value of $P_3$ and the 480 hr running time, and corresponds to
281: $0.6\,$mHz. This rate of two PMT signals in coincidence is consistent
282: with the observed rate of single PMT signals above threshold ($\sim
283: 6\,$Hz) and our $20\,\mu$s timing window. For comparison, the
284: fitted rate of muon decays in the scintillator is $17\,$mHz ($\sim
285: 1\,\text{min}^{-1}$).
286:
287:
288: From the charge averaged lifetime of muons in what is essentially a
289: carbon target (the ratio of hydrogen to carbon in plastic scintillator
290: is ~1:1), and the lifetime of $\mu^-$ in carbon\cite{reiter}, it is
291: straightforward\cite{rossi} to measure the charge ratio abundance
292: $\rho=N(\mu^+)/N(\mu^-)$ of low-energy ($E\lesssim 200\,$MeV) muons at
293: sea-level. For example, from our measured lifetime, we find $\rho=
294: 1.08\pm0.01$ (statistical error only), averaged over the angular
295: acceptance of the detector, a value consistent with the diminishing
296: trend\cite{inpc} for $\rho$ as the muon momentum approaches zero.
297:
298: \section{Demonstration of Relativistic Time Dilation}
299:
300: The stopping rate of muons in the detector as a function of altitude
301: above sea level can be used to demonstrate relativistic time dilation.
302: Although the detector design is non-optimal for this demonstration
303: since it is sensitive to muons with a range of velocities as well as
304: non-vertical trajectories, it has the advantage that no bulky velocity
305: selecting absorbers or additional trajectory defining
306: scintillators\cite{easwar} are required. The idea is simple. The total
307: number of stopped muons in the detector in some fixed time interval
308: and at some fixed altitude above sea level (a.s.l.) is measured from
309: the decay time time histogram. A lower altitude is selected and
310: predictions made for the new stopping rate that do and do not include
311: the time dilation effect of special relativity. Measurement then
312: discriminates between the two predictions.
313:
314: To make the comparison between the competing assumptions meaningful,
315: the effects of the energy loss of a muon as it descends in the
316: atmosphere as well as the shape of the sub-GeV/c muon momentum
317: spectrum\cite{muspec} near sea-level should be included. The first
318: effect tends to increase the transit time of the muon from one
319: altitude to another and the other tends to over emphasize the
320: effects of time dilation.
321:
322: The transit time $t^{\prime}$ measured in the muon's rest frame as it
323: descends vertically in the atmosphere from a height $H$ down to
324: sea-level is given by
325:
326: \begin{equation}
327: t^{\prime}=\int^0_H {\frac {dh}{c\beta(h)\gamma(h)}}\label{time}
328: \end{equation}
329:
330: \noindent where $\beta$ and $\gamma$ have their normal relativistic
331: meanings, $dh$ is a differential element of pathlength and $c$ is the
332: speed of light. All quantities on the right-hand side of
333: Eq.~(\ref{time}) are measured in the detector rest frame. As the muon
334: descends it loses energy in a manner described by the Bethe-Bloch
335: equation\cite{leo} so the integral can be evaluated numerically if
336: great precision is desired. Instead, we use the common approximation
337: that a singly-charged relativistic particle loses energy by ionization
338: and excitation of the medium it traverses with a magnitude $dE/dx=
339: 2\,{\text {MeV}\,\text{g}^{-1}\text{cm}{}^2}\,(\equiv\! S_0)$.
340: Eq.~(\ref{time}) becomes
341:
342: \begin{equation}
343: t^{\prime} \simeq {\frac{mc} {\rho S_0}}\int^{\gamma_1}_{\gamma_2}{\frac{d\gamma}
344: {\sqrt{\gamma^2 -1}}}\label{time2}
345: \end{equation}
346:
347: \noindent Here, $\gamma_1$ is the muon's Lorentz factor at height $H$,
348: $\gamma_2$ is its Lorentz factor just before it enters the sea level
349: scintillator, $m$ is the muon mass and $\rho$ denotes the
350: pathlength-averaged mass density of the atmosphere. We take
351: $\gamma_2\simeq 1.5$ since we want muons that stop in the scintillator
352: and assume, consistent with our detector geometry, that stopped muons
353: travel an average distance $s=10\, \text {g/cm}^2$ in the scintillator
354: . (See muon range-momentum graphs from the Particle Data
355: Group\cite{pdg} for correlating a muon's range with its momentum.) The
356: appropriate value of $\gamma_1$ depends on the height $H$ where we
357: take our upper measurement and is computed from the energy $E_1$ a
358: muon has at that height if it is to arrive at the sea-level detector
359: with $\gamma_2=1.5$ (corresponding to energy $E_2=160\,$MeV). Clearly,
360: if a muon loses energy $\Delta E$ in traversing a vertical distance
361: $H$, then $E_1 = \Delta E + 160\,$MeV. The quantity $\Delta E$ can be
362: computed from the Bethe-Bloch equation or estimated from the above
363: rule-of-thumb for minimum ionizing particles and properties of the
364: standard atmosphere.
365:
366: Since the time dilation demonstration relies on stopping muons in the
367: detector, we must account for the fact that muons that eventually stop
368: in the lower detector have, at the position of the upper detector, an
369: energy that is greater than those muons that would be stopped in the
370: upper detector. Since the momentum spectrum of sub-GeV muons near
371: sea-level is not flat, but peaks at muon momentum
372: $p_{\mu} \sim 500\,$MeV/c, we correct for this affect so that the
373: effective flux of incident muons is appropriately normalized. (This is
374: easy to see if you assume muons don't decay at all and only lose
375: energy in the atmosphere as they descend.) We do this by measuring the
376: ratio of stopping rates at a pair of altitudes to determine a single
377: scaling factor that we can apply to other pairs of altitudes.
378:
379:
380: To illustrate the procedure, we measure the muon stopping rate at two
381: different elevations ($\Delta h=3,000\,$m between Taos, NM and Dallas,
382: TX) and compute the ratio $\text R_{\text {obs}}$ of observed stopping
383: rates ($\text R_{\text{obs}}(\text {Dallas}/\text{Taos})=
384: 0.41\pm0.05$.) The transit time $t^{\prime}$ in the muon's rest frame
385: for vertical trajectories between the two elevations is computed using
386: Eq.~\ref{time2} and yields $t^{\prime}=1.32\tau_{\mu}$. This
387: corresponds to a naive theoretical stopping rate ratio
388: $R=\exp(-t^{\prime}/\tau_{\mu})=0.267$. The double ratio $R_0 =
389: R_{\text{obs}}/R =1.5\pm 0.2$ is then interpreted as a correction
390: factor for the shape of the muon momentum spectrum. Note that this
391: same correction tends to account for muons with non-vertical
392: trajectories that stop in the detector. These slanting muons with a
393: projection onto the vertical axis of distance $H$ travel further in
394: the atmosphere and hence start with more energy than their purely
395: vertical counterparts.
396:
397:
398: To verify that the procedure is sensible, we choose a new elevation
399: ($H=2133\,$m a.s.l. at Los Alamos, NM), compute the transit time
400: $t^{\prime}=1.06\tau_{\mu}$ and the expected stopping rate ratio
401: (Dallas/Los Alamos) $R_{\text {thy}} =
402: R_0\exp(-t^{\prime}/\tau_{\mu})=0.52\pm 0.06$. The observed ratio
403: $R_{\text obs}=0.56\pm0.01$, showing good agreement.
404: Table~\ref{tab:stopdata} summarizes relevant measurements and lists in
405: the third column calculated proper transit times for vertical muon
406: trajectories, in units of the proper muon lifetime $\tau_\mu$ and
407: relative to Dallas' elevation.
408:
409:
410: \begin{table}
411: \caption{\label{tab:stopdata}Muon Stopping Rate Measurements and
412: Calculated Proper Transit Times}
413: \begin{ruledtabular}
414: \begin{tabular}{lcc}
415: Elevation (meters a.s.l.)& Observed Stopping Rate (min${}^{-1}$)& Proper transit time ($\tau_{\mu}$)\\
416: 146 & $1.24\pm0.01$ & $-$\\
417: 2133 & $2.21\pm0.05$ & $1.06$\\
418: 3154 & $3.00\pm0.34$ & $1.32$\\
419: \end{tabular}
420: \end{ruledtabular}
421: \end{table}
422:
423:
424: To compare the stopping rate measurements with the competing
425: assumption that there is no time dilation effect (``ntd''), we proceed
426: as before except we calculate the transit time in the detector rest
427: frame and we assume all muons travel at the speed of light so as to
428: {\it minimize} the effect of time dilation. For the case of transit
429: between Los Alamos and Dallas, the transit time $t_{\text {ntd}}$ in
430: the detector rest frame is $t_{\text {ntd}}=6.62\,\mu$s, implying an
431: expected stopping rate ratio $R_{\text {ntd}}=R_0\exp(-t/\tau)
432: =0.08\pm 0.01$, a result strongly disfavored by observation.
433:
434:
435:
436: \section{Summary}
437:
438: We have designed a compact and low-cost apparatus for measuring the
439: mean muon lifetime and for demonstrating relativistic time dilation
440: suitable for undergraduate teaching. An electronics schematic and
441: Tcl/Tk data acquisition/display software are available upon request.
442:
443:
444: \begin{acknowledgments}
445: The technical assistance of H. van Hecke and L. Lu is greatly
446: appreciated, as well as the financial support of the Lightner-Sams
447: foundation.
448: \end{acknowledgments}
449:
450: %\subsection{\label{sec:level2}Second-level heading: Formatting}
451:
452: %\subsubsection{\label{sec:level3}Third-level heading: References and Footnotes}
453:
454: \bibliography{muon}% Produces the bibliography via BibTeX.
455:
456: \end{document}
457: