THE UNIVERSITY OF MICHIGAN COLLEGE OF LITERATURE, SCIENCE, AND THE ARTS Department of Physics Technical Report No. 29 A SEARCH FOR MASSIVE PARTICLES IN COSMIC RAYS L. W. Jones, D. E. Lyon, Jr., P. V. Ramana Murthy, G. DeMeester University of Michigan, Ann Arbor, Michigan Ro W. Hartung, S. Mikamo, Do Reeder, A. Subramanian University of Wisconsin, Madison, Wisconsin Bruce Cork Lawrence Radiation Laboratory, Berkeley, California B. Dayton California State College, Los Angeles, California A. Benvenuti, E. Marquit University of Minnesota, Minneapolis, Minnesota P. D. Kearney Colorado State University, Fort Collins, Colorado A. E. Bussian High AltitucdeObservatory, Boulder, Colorado F. Mills, C. Radmer, W. R. Winter Midwestern Universities Research Association, Stoughton, Wisconsin ORA Pr-ojtect 03106 ORA Project 08840+ under contract with: DEPARTMENT OF THE NAVY OFFICE OF NAVAL RESEARCH WASHINGTON, D.C. CONTRACT NO. Nonr-1224(23) +NATIONAL SCIENCE FOUNDATION WASHINGTON, D C. CONTRACT NO. GT-7273 administered through OFFICE OF RESEARCH ADMINISTRATION ANN ARBOR July 1967 Distribution of this document is unlimited.

This report was prepared as a paper for submission to Physical Review.

ABSTRACT A search for heavy particles with a mass greater than that of a nucleon, the existence of which has been suggested by higher symmetry schemes, was performed with an apparatus set up at Echo Lake, Colorado (elevation 10,600 ft). The search was sensitive to strongly interacting particles with masses in the range 5-15 BeV, with no restriction imposed on their electric charge. The method used was to measure the time interval between the arrival of strongly interacting particles and accompanying air shower particles. This information, coupled with information from a measurement of the particle's energy and range of absorption in a total absorption spectrometer, enabled a distinction to be made between massive elementary particles, nucleons, and nuclei. In an operating period of 1542 hours and with an aperture of 0.78 m2 sr, one delayed event was found whose behavior in the total absorption spectrometer was atypical of a nucleon or nucleus. If one considers this event to represent the arrival of a massive particle then its mass, calculated assuming that its production occurred 1 km above the apparatus, is approximately 6.5 BeV. This one event corresponds to a flux of the order of 1010 (cm sec sr) 1 where a correction for detection efficiency has been included. -1 -

I. INTRODUCTION Following the success of SU(3) and SU(6) symmetry schemes in the classification of known hadrons, theoretical speculations have arisen concerning the possible existence of fundamental 1-6 subunits of particles. In these speculations hadrons are considered to be composites of these hitherto undiscovered fundamental subunits, and in the simplest schemes1'2 there is a triplet of them, called "quarks," where the elements of the triplet have fractional charges. In other versions there are more than three elements, and the charges carried by these are 3-6 5 integral.36 Theoretical conclusions suggest that the masses of these fundamental elements are likely to be of the order of 10 BeV. tfheree have 'been several searches that used accelerators or 0cos1mic ays A c t i f ion1 ly ch ar e d p a rt i c 1 e s 7E- a d a e 1-ew o i d SfaL c ti o ally o it e Cat 18 l c a —, E d eL ct1 e 1 - 22 cother searche for naturally occurring stable quarks have been I II I 8 I | I performed that used mass spectrometric, optical spectroscopic, and other methods.2328 The results of these searches were negative and have thus set lower limits for the masses of these fundamental subunits, of the order of 5 BeV, and corresponding upper limits for their production cross-section. Calculations have been performed concerning the production crosssection for these heavy particles in nucleon-nucleon collisions,2930 and an estimate based on the peripheral model predicts a crosssection of the order of 1 bb/nucleon for a particle with a mass -2 -

-3 -of 6 BeV. The method adopted in the present search is similar to the one suggested by Damgaard et al.31 and consists of measuring the time interval between the arrival at the apparatus of strongly interacting particles and their accompanying air showers. The distribution in the time delayed arrival of a particular strongly interacting particle with respect to the associated air shower arrival is a measure of the mass of the particle and will be explained in the next section. The merits and defects of the technique are also discussed there. II. THE METHOD A particle of mass M and energy E produced in the atmosphere in a collision at a height Z above the apparatus will arrive at the apparatus delayed in time with respect to the arrival of the associated air shower of extremely relativistic particles (muons, electrons, etc.) by a time t = (YZ) = z Y ~ 1, (1) 2y c where y = E/M and c is the velocity of light. If one assumes a Z of 1 km, corresponding to about one nuclear interaction mean-free path in the atmosphere at 10,000 ft, and a y of ten, then Equation (1) gives t = 17 ns as the order of magnitude of time delays that can be expected. Also, in the production

-4 -of massive particles with mass M, the mean energy, Ev, of the particles when produced in pairs in nucleon-nucleon collisions near the threshold energy, Eth, is Ev Eth 1 (2M)2 M2 M le n Thus, with M = 10 BeV, one finds that Eav is of the order of 100 BeV, so that one has to use a device for detecting and measuring particles with energies of the order of 100 BeVo The distribution in arrival times for these massive particles depends on the intensity, ml(x), of hadrons at various heights in the atmosphere, and the density distribution, ~2(Z), of the atmosphere itself, as well as the survival probability, <o(x), for the massive particles to reach the apparatus from a height x above it. Here x is measured in gm/cm2 and Z in kilometerso To a first approximation one can consider ml(X) cc eX/ka and 3(x) cc e x/h so that

-5 -m(Z,x) 2(Z) e a For the case where Xh = Xa is a function of Z alone and is m(Z) c m2(Z) If one assumes an exponential variation of atmospheric pressure with Z, then $(Z) c eZ/Zo where Z~ is the scale height of the atmosphere, taken to be 7 km. This distribution upon transformation from Z to t via Equation (1) becomes m(Y, t) c et/to(Y), (2) the distribution in arrival times of massive particles with y = E/M. Here to(Y) is the mean arrival time and is given by to(Y) 2= (3) 2y c The distribution ((y, t) was calculated for particles with a mass of 10 BeV and an energy of 100 BeV for three choices of kh5 and the results are shown in Figure 1. The actual distributions will be somewhat steeper because of the poor detection efficiency for air showers with an origin high in the atmosphere. There have been efforts to calculate the distribution -(y, t) in greater detail,32'33 and, while the results are qualitatively

-6 -similar to Equation (2), they are not directly applicable to this experiment. With these results, then, a measurement of t and E leads to the mass of the particle via the relation 1 2 2 2E, ct \ M =- (^-00) (4) with contributions to the error in the determination of the mass given by AM _ E to M E 2t (5) The method imposes certain requirements that the hypothetical particles must satisfy in order that they can be detected. First, they must have a mean lifetime that is greater than about 10 seconds so that they can reach the apparatus at all, Second, they must lose a significant fraction (> 5% per interaction) of their energies through nuclear interactions so that they can be detected in the total absorption spectrometer used in this experiment to measure event energy. And, third, they should acquire only about the same order of magnitude of the transverse momentum that is acquired by nucleons in nuclear collisions, so that the massive particles are not greatly displaced spatially from their associated air shower when they reach the apparatus. In the present method, the search for massive particles could be made without reference to their electric charge. However, this made it necessary to use other methods to distin

-7 -guish massive particles from the nuclei of elements like helium and carbon that are present among cosmic ray primaries and which have a small but non-negligible probability of 'being observed at mountain altitudes. III. EXPERIMENTAL ARRANGEMENT The arrangement of the apparatus used to search for massive elementary particles is described below and is shown in Figure 2. The central element of the apparatus was a total absorption spectrometer which was used to measure the energy of nuclear active particles incident on it. The spectrometer was fabricated 2 from an iron stack with a thickness of 1070 gm/cm and an area of 3' x 6', with plastic scintillators 3/4" thick as probes placed at seven levels in the absorber material, Each of the six uppermost scintillators was viewed by four photomultiplier tubes so that there was a nearly uniform response for varying positions of the passage of ionizing particles. The bottom scintillator was 4' x 8' in area and was viewed by six photomultiplier tubes. The scintillators were calibrated in terms of the energy loss of cosmic ray muons passing through them, so that the level of ionization recorded by a scintillation counter could be turned into the level due to an equivalent number of relativistic muons, that lose an energy in iron of 1.7 MeV (gm/cm2). A minimum level in the sum of pulses from the scintillators served to select triggers that corresponded to hadron energies of the order of or greater than 10 BeVo

-8 -The spectrometer was flanked by several scintillation counters to detect accompanying air showers. The total area presented by these air shower counters, which was 130 ft, was divided into four groups of counters of approximately equal area. A typical counter in each group consisted of a 2' x 2' x 2" plastic scintillator viewed by one photomultiplier tube. Additional elements of the apparatus included a wide gap spark chamber and a system of six gas proportional counters. In cases of interest the spark chamber provided the direction and number of charged particles incident on the spectrometer, and in cases of a single incident charged particle the gas proportional counters yielded the chargeo Here, the level of ionization corresponding to unit charge was determined from the level recorded when relativistic muons passed through the counter systemo The signals from the seven counters in the total absorption spectrometer were logarithmically digitized, and the signals from the six gas proportional counters were linearly digitized into seven binary 'bits of information per counter, The time delay between the spectrometer trigger signal and the signal from each of the four air shower counter groups, measured with respect to the spectrometer signal, was digitized through a system of time to height converters and an analog to digital circuit. All digital information was then recorded on punched paper tape. The dynamic recording ranges for the various elements of the apparatus were, 1 to 1000 minimum

-9 -ionizing muons, -200 ns to +200 ns, and 0 to twice minimum ionization, for the spectrometer, time interval, and gas proportional counter signals, respectively. The apparatus was housed in a wooden building, and the associated electronics was located in an adjacent trailer. The equipment was operated at Echo Lake, Colorado (elevation 10,600 ft), during the fall and winter of 1966-67. IV. PERFORMANCE OF THE APPARATUS It should be noted at the outset that the time distributions presented herein are from the point of view of the shower counters so that massive particle-like events appear with t < 0. A. Time Resolution Errors in the measurement of the time interval between a signal from the spectrometer and a signal from the air shower counters arose from (a) coincidence and discriminator jitter, (b) electron transit time jitter in the phototubes, (c) transit time variations of light in the counters, and (d) the spread in the arrival time of particles belonging to the air shower. The weights of these effects in determining the time resolution of the apparatus were measured and the results summarized in Table I(a). The conclusion is that the time resolution of the apparatus was approximately 16 ns and that this was mainly due to the spread in time of the arrival of the air shower particleso This resolution is to be compared to the experimental time resolution, that has been taken to be given by the full width

-10 -at half maximum of the accompanying air shower time distribution plotted using events whose energies, as measured in the spectrometer, were greater than 100 BeV. This experimental distribution is shown in Figure 3. The distribution shown in Figure 3 is a mixture of events in which either two, three, or four shower counter groups gave signals, all within a time spread of 40 nso What was plotted, then, is the average time "delay" of the shower counters with respect to the spectrometer, averaged over two, three, or four shower counter groups. The half widths for the separate distributions contributing to Figure 3 are given in Table I(b) and are to be compared to the expected half width of r/2 = 8 ns The broadest of these contributing distributions is the distribution for events where only two shower counter signals were within the 40 ns allowed time spread, and the widths naturally decrease as the averaging is done over more shower counter groups. The asymmetry of Figure 3 seems to stem from the fact that signals from the spectrometer were used in conjunction with signals from the air shower counters since in the studies made to obtain the values for Table I(a) no asymmetric distribution was observed. An interpretation of the asymmetry is that the shower counters were triggered by slow neutrons (_ 200 MeV/c) or other backward-going particles from the nuclear cascade in the spectrometer. The relative frequency of such events, though, as judged from the size of the asymmetry of Figure 3, was only of the order of 3 percento Also, the asymmetry

-11 -becomes negligible if only those events are included in the plot where shower counter averaging was done over three or four groups. This can be interpreted as an effect of averaging or as a real effect stemming from the relatively lower probability that back-splash from the spectrometer would give signals in three or four shower counter groups. B. Energy Resolution The energy of each recorded event was calculated using the relation f [r N N. 1+N. + Ec + - xi + AN. (6) i=2 Here, the N's are the equivalent number of particles with specific ionization i, the x's are the thicknesses of the absorber between adjacent counters, and f, taken to be 1.3. is a factor to account for unsampled energy losses in the absorber. Also, it was assumed that experimental absorption prevailed at large thicknesses so that the last term in Equation (6) represents an extrapolated area of the shower curve beyond the fifth probe: eX/AN5dx = AN5 J 5 5 o The absorption mean-free path, A, was either calculated from N = Ne 533/A, when possible, or an average value of A = 200 gm/cm was used. The reason for not including the count N6 in the calculation of energy is that the output from this

-12 -counter was found to be erratic. The error in the estimate of energy had two sourceso (1) error in integration of the ionization curve and (2) fluctuations in the division of the incident energy into sampled and unsampled energy losses. The energy resolution that resulted 34 from these sources of error, as discussed in a separate paper4 was ~20%, which is small compared to the mass ratio of massive particles to nucleons that this experiment is seeking to distinguish (see Equation (5)). However, it should 'be emphasized that since massive particle events are expected to be rare, the validity of their existence is related to the tail of the energy resolution function rather than to its width. That is, one must estimate the probability that nucleons with an energy of a few BeV, for example, could give a spectrometer cascade characteristic of cascades where E_ 0 30 BeVo Also, if the massive particles dissipate a smaller fraction of their energies in the spectrometer than do nucleons, the effective energy resolution for them deteriorates because their calculated energy then drops into a range where they may not be distinguished from lower energy nucleonso This is discussed below in evalu. ating the significance of the results obtainedo V. DATA COLLECTION AND ANALYSIS Ao Data Collection The collection rate for events whose calculated energy was greater than 10 BeV and that had an accompanying air shower

-13 -was 4 per minute. Here, the 10 BeV trigger threshold level was determined by summing the signals from counters two through six of the spectrometer, counting from the top, and discriminating the summed pulse for an ionization level corresponding to the equivalent of 30 or greater minimum ionizing muons (counter one was excluded from the trigger threshold sum in order to prevent triggering on air showers). The "definition" of an accompanying air shower was that signals from the air shower counter groups should occur within =200 ns of the spectrometer signal and that at least two of these should be within 40 ns of each other. The effective area times solid angle of the spectrometer was 0.78 m~ sr, where the calculation includes a small correction for the zenith angle dependence of the nucleon flux in the atmosphere. During 1542 hours of operation 3 x 10 events were collected. oB Data Analysis To examine the behavior of the events collected a three parameter distribution of the observable characteristics of an event was studiedo The parameters used were t, the hadron arrival time with respect to its accompanying air shower arrival, Ec, the calculated energy of the event, and n, the number of counters in the spectrometer that registered an ionization level greater than that due to the passage of one minimum ionizing muon. This last variable, n, is a measure of the range of the secondaries in the spectrometer and is related to the true energy of the hadron responsible for the event.

-14 -The distribution in these three parameters can be represented by -t/t (yt) P(En,t) = G(Ecn,Et) e S(Et) dEt where Et is the true energy, Yt is Et/M, S(Et) is the energy spectrum of the detected hadrons, and G is a function that gives the correlations between calculated energy, range of secondaries, and true energy. Also, it was assumed that the calculations producing Equation (2) holdo For a given Ec n, and M, then, and with a Gaussian-like correlation function, the distribution in the variable t is mainly exponentialo The distribution in t for events whose E was of the c order of or greater than 10 BeV, and for all n, is shown in Figure 4. Here, a large tail is observed in the region t < 0, which is not due to chance coincidenceso Chance coincidences should produce a flat background in the sensitive window width displayed, and they should only occur at the rate of 4 events per 105 triggers, as judged from the rate of air shower signals obtained with this apparatus. The distribution in E for these delayed events3 which was found to be independent of the delay, has a mean value of about 10 BeV and is shown in Figure 5. The mean value of 10 BeV makes it unlikely that lighter particles like kaons or pions contribute to the events in the delayed spectrum0 In fact, a calculation of the mass of the particles responsible for the delayed spectrum, using Equation (4) with to taken from Figure 4 and E = 10 BeV, gave the value 0O7 BeV9 so that

-15 -the majority of the delayed events give a distribution that is at least consistent with a low energy nucleon distribution. The projection of P(Ec n, t) onto the n axis is shown in Figure 6 for five energy bins. Note that this distribution in n for events that are predominantly "prompt" events and thus typical of nucleons is peaked strongly at n = 1-2 in the E < 10 BeV plot, and that the distribution shifts so that C in the Ec > 100 BeV plot the peak is at n = 6. This makes it very probable that low energy nucleons (Ec < 10 BeV) will give signals usually in only one or two counters so that the correlation between n and Ec can be used to evaluate the "reality" of energetic, delayed events. Figure 7 shows the n distribution for events where t < -30 ns and for two energy ranges. Here the main distribution in each plot is representative of the distribution for low energy nucleons so that for these it is probably true that Et << Eco Since the delayed events are mainly of the type n = 1 or 2 and their t distribution is consistent with that of low energy nucleons, we believe that they are virtually all nucleons with energies less than 10 BeV appearing to have greater energies, and that this arises because a nucleon of only a few BeV can give anomalously large pulses in one or two counters from nuclear stars in or close to the scintillators. There was one event in the delayed spectrum (t < -30 ns), however, whose calculated energy was 36 BeV and whose delay was -45 ns, that registered greater than minimum ionization levels

-16 -in six counters. That this one event which appears in the E > 30 BeV plot of Figure 7 shows n = 6, and that it is separate from the main distribution there is taken to mean that for this event Et may be close to Eco In fact, from the frequency distribution of n for events with Ec < 10 BeV, shown in Figure 6, the probability for such events to be of the type n = 6 was calculated and the result was 0.34%, With a total of 6 events on the E > 30 BeV plot of Figure 7, then, the expected probability of observing one such event in this experiment is about 2%. Also, the expected probability of observing one event due to chance coincidence between a delayed air shower and a spectrometer signal of the type Ec > 30 BeV and n = 6, in the sample of 3 x 105 events, is 6o. Thus, the delay of the energetic, n = 6 event becomes meaningful. For this one interesting delayed event, all four shower counter groups gave signals within 5 ns of each other, and the proportional counters gave signals which saturated the ADC circuits. Table II(a) lists the available information on the event. If it is assumed that the apparatus had unit efficiency for the detection of such particles, then, this one event -11 2 -1 corresponds to a flux of 2.3 x 10 (cm sec sr)- Also, the mass of the hadron responsible for the event, calculated using Equation (4) and assuming that the origin of the particle was 1 km above the apparatus, is 6.5 BeV.

-17 -VI. CONCLUSIONS Heavy nuclei exist among the cosmic ray primaries and there is the possibility that these nuclei could be responsible for the one interesting event. For example, the flux of a-particles with an energy per nucleon of 10 BeV that reach the depth in the atmosphere at which the apparatus is located -8 2 -1 is estimated to be of the order of 10 (cm sec sr) o These could give time delayed signals of the order of -45 ns thus simulating massive particle-like events. However, while it is the total detectable energy, Ec, that is measured in the spectrometer, the range of the nuclear cascade that develops depends on Ec/A, where A is the mass number. Therefore, the n distribution for nuclei with an energy Ec is expected to be peaked more toward low n values than the n distribution for nucleons with the same E. There are three delayed events in our data, other than the interesting one, with calculated energies of about 100 BeV. Details concerning these other events are listed in Table II(b). They are all n = 2 events and are listed as examples of what one expects of heavy nuclei. In our judgment, therefore, the n = 6 event mentioned above as a candidate for a massive particle-like event exhibits a nuclear cascade curve that is not characteristic of the behavior expected from heavy nuclei, and, thus, cannot be attributed to such sources. To support this conclusion, we point out that events giving delays > 30 ns would have to be produced at least 2 km above the apparatus, and with the true

-18 -energy per nucleon limited to about 10 BeV it is doubtful whether accompanying showers of detectable size would be generated at all, even by iron nuclei. Also, from the discussion in the preceding section, it seems unlikely that the interesting event is due to a very low energy nucleon where Ec >> Et, or due to the chance coincidence of a delayed air shower with the Ec > 30 BeV and n = 6 spectrometer signal. However, since one event cannot be construed to be evidence for the existence of the hypothetical particles sought for, we tend to favor the interpretation that this one unusual event is a nucleon in spite of the total probability for observing one such event in this experiment of only 8%, and merely use the event to quote upper limits for the flux and cross-section of fractionally or integrally charged massive particleso A calculation of the total and detectable flux of massive particles, presented in the Appendix to this paper, was performed, and the results are listed in Table III, for the choices kh = Xa and Xh = c It is the calculated detectable flux that was compared with the observed flux of 2o3 x 10 (cm2 sec sr)-1 to obtain a value for the cross-section for massive particle production as a function of the mass. The resulting sensitivity of this work is shown in Figure 8, and our upper limit to the massive particle flux is compared to those obtained in other experiments in Table IVo

-19 -APPENDIX: FLUX AND CROSS-SECTION The calculation of the flux of massive particles with an energy E at a depth X, in gm/cm2 in the atmosphere, is developed below and is based on a model developed earlier by some of us. In the model we assume that massive particles of mass M are produced in pairs in nucleon-nucleon collisions and that they are produced at rest in the c.m. system of the interacting nucleons, with an energy independent cross-section, aHN that above threshold is equal to r('n/Mc). Other calculations of the flux of massive particles based on similar models have been performed and exist in the literature 32o35 A. Intensity Distribution The source spectrum of the massive particles (i.e., the number of massive particles produced in 1 gm/cm2 at a depth y, with a mass M and in the energy range E to E + dE) can be written as S(EyM) = J N(E',y) W(EE',M) dE'/Xin min Here Xin is the inelastic nucleon collision mean-free path in the atmosphere, N(E',y) is the nucleon intensity distribution, and W(E,E',M) is the differential production spectrum for massive particles of mass M produced with an energy E in a nucleon-nucleon collision where the incident nucleon energy was E'. The depth in the atmosphere is to be measured from

-20 -the top in gm/cm2, and energies are to be measured in BeV. Also, the nucleon intensity distribution is given by N(E1,y) = K E-(c+1) eY/ a where the form of N and the values of the constants appearing are taken from experiment and are K = 1, and C = 1.7o The absorption mean-free path in the atmosphere for nucleons, denoted by a' has the value of 120 gm/cm2 In nucleon-nucleon collisions the gamma, Ycm' of the com. system is given by Y' m E'/2Mn, if E' >> Mn and where M denotes the mass of the nucleon, so that if massive particles are to be produced at rest in this system the laboratory energy, E, of each must be related to the incident nucleon energy, E', 'by E = M(E/2M )2 The differential production spectrum then has the form 1 W(EE',M) = 2(aH/ain) 6[M(E/2Mn)2 - E] where the 2 represents the fact that the massive particles are to be produced in pairs, and ain is the inelastic nucleon collision cross-section corresponding to Xin. Then, for nucleon energies above the threshold for the pair production of massive particles, given by Eth = 2M2/M and corresponding to a minimum laboratory energy for each massive particle of Emin o M2/Mn the source spectrum becomes

-21 -S(Ey,M) = 4K/XH (2Mn/M2)JC E-(2c+) e a where we have put Xin gin/OH = H' The intensity distribution for massive particles, assuming for them an absorption mean-free path in air of Xh, is thus x _ (X-y)/x H(E,X,M) = J S(E,y,M) e h dy 0o For the choices Xh = Xa' and Xh =, corresponding to the assumptions that massive particles attenuate in the atmosphere like nucleons, and that they do not attenuate at all, respectively, the H distributions are H a(EXM) = 4K/H (2Mn/M2)- E (26 )X e a and XHh (EXM) = 4K /H (2M /M2) ' E (2E+ 1)le X/ n J Integration over E then gives the expected flux of massive particles with energies greater than the minimum possible energy, at the depth X in the atmosphereo That is, xh=~X( X X)(Xa( H (XM) = H (EXM) dE 2 M2/M and X kh=c Xh=CO H (XM) H (E,X,M) dE M2/M n Table III lists the expected values of H at the depth in the 2 atmosphere X = 715 gm/cm, the level of observation in this experiment, for various choices of the mass of the particleso

-22 -B. Detectable Intensity The detectable flux will be less than the expected values because of the finite inefficiency of the apparatus for the detection of massive particles. This inefficiency arises from two sources: one, the associated air shower detection efficiency that results from the finite areas used for shower detection, and, two, the massive particle detection efficiency that results from the restricted time window width for delays present in the apparatus. That is, one must evaluate both the probability that the associated air shower is detected in our apparatus and the probability that the time delayed arrival of the massive particle is within the sensitive window width of -40 ns to -130 ns (the range t < -130 ns was excluded because of noise problems with the time-to-height circuits). The shower detection efficiency was calculated by expanding the nucleon intensity distribution as -y/X. n ne N(E'y) = N(E',0) e in (- ) n n=o (in where Xa = in/(l-r)) and i is the nucleon inelasticity. Mutually consistent values for the constants appearing are Xa = 120 gm/cm2, Xn = 83 gm/cm2, and r = 0.5. The th term Nj. in this expansion represents the number of nucleons arriving at the depth y that have suffered j collisions in the atmosphere in degrading to the energy E'. Thus, for each term one can calculate the energy transferred to the air shower and the corresponding probability for detection of it,

-23 -Assuming, then, that the energy left over after production of the massive particle pair also contributes to air shower production, the energy transferred to the air shower after j collisions can 'be written as Ej - E' 2E EJ 2E sh j where E'/r is the laboratory energy of the incident primary cosmic ray nucleon, E' is the energy of the incident cosmic ray nucleon (after j collisions in the atmosphere), and E is the laboratory energy of each massive particle. Assuming that the number of electrons in the shower at the observation level is given by E(h)/ 2, the shower size at cascade maximum, one can find the density of shower particles at the shower detectors from Aj(r) E() f(r)/2R12 where f(r) is the Nishimura-Kamata lateral distribution function with age parameter unity and Ri is the so-called scattering length36 If the area of each shower detector is S, then with four detectors in our apparatus -4SA.- -3 -SA. Pj(E') - 1 - e -4e J (I - e ) represents the corresponding probability that two or more shower counters will "fire," and is the shower detection efficiency for nucleons that have j collisions before producing a massive particle pair. The detectable flux of massive particles at the depth X ca n then be written as

-24 -co X -y HD(X,M) = X dE Pt(E,M) I dy e ne h M /Mn 00 ""1 co n ne / 2 X dE'N(E,O) n ( ) Pn(E) 6 [M(2 ) -E] E nn=l n Here, we have neglected the contributions to the air shower from the massive particles, as they traverse the remaining distance from their point of production to the apparatus. Also, Pt(EM) is the probability that the massive particle will arrive at the apparatus with a delay that is within the time window and is given by Xh=X -40/to(Y) -130/t (Y) PI. ha(EM) = e - e or 130 [~-t/t~(Y) J h130 d -t/to(y) + X(l-e / 0)/Xa Xh=- M 4dt e P (E5M) co, J dt el[ o where to(y) is given by Equation (3)o It is useful to note that a calculation was performed using the method outlined above to estimate the detectable shower-associated nucleon fluxo The ratio of the detectable shower-associated flux to the total expected nucleon flux was compared to the experimentally observed ratio and found to

-25 -agree to within 5%. This gives one some confidence in the method, particularly in the use of the Nishimura-Kamata lateral distribution function. Further, an experimental determination of SA was made using the observed ratio of the frequency of the events in which any two shower counter groups recorded counts to the frequency where any three groups recorded counts. This ratio varied from 1:1 at an event energy of about 100 BeV to 1:2.5 at 500 BeV. Thus, from the relation R = 3e-SA/2(1-eSA) the values of SA were found to vary from 1 to 2. This means that on the average less than 18% of the nucleon shower-associated flux present goes undetected by not setting off at least two shower counter groups.

Acknowledgments It is a pleasure to express our appreciation to Professor Mario Iona, Mr, Robert Venuti, and the University of Denver for their hospitality and cooperation with us in the use of the Echo Lake High Altitude Laboratoryo We would also like to thank Mr. R. T. Brown of M.U.R.A. for his vigilant operation and maintenance of the equipment. One author (P.V.R.M.) would like to acknowledge the support of the University of Michigan Institute of Science and Technology. We would like to thank the National Bureau of Standards, Boulder, for making available to us their computing facility. Also, D. Lyon and A. Subramanian would like to thank M. L. Good for many helpful discussions regarding the experiment and the analysis of the data, We would also like to thank T, Novey at Argonne National Laboratory for the use of the plastic scintillator used in the calorimeter.

Table I(a): Width of Time Distribution Due to Various Sources Cause of Spread r, ns Circuitry 0.8 Electron time jitter between two phototubes 5.0 Light transit in scintillator 4.4 Shower front 10.0 Table I(b): Width of Time Distribution Actually Observed to be Compared to an Expected Half Width r/2 = 8 ns Width of Left Half Case of Signal From Width of Left Half of the Distribution One shower counter 9.0 Two shower counters 7o7 Three shower counters 7.3 Four shower counters 5.9

Ln CH a- ~ c~ ~ ~ ~ ~ ~ *o H qOQOO * HLCH.-Q0 co n Ln H O Lf) Ln C\JLC *H Ln 'k O >O O H HC\J i eqn ' a l HHHH Hr 'H 00 S C c O O O O O O * 0 * l oe O>J\ COH 0 > OH Ln f c * n 0 a o H CN o I C \ I HI I r-!.-i CH CO r-H o o o o o H* * C\i J HQr - o c mc \- Lc l - oj m rJO e* *o L\ Q\C\J OOLH 0 *H e'l -p I.. 0......O..J 0 19 LN H CJ Cr)r-1f Lnf O r CH -0 r -4 1-) LH eD H CO I) f I I I S i0! I I I-H l r r'c _ z_ ___0 _-_ c3 ~l 0 U) t - U) H ---- -— 0 U -- -------- Ho eH -P, H >, -H -H3 a) CJ 3 r C C D oH CX ~ 0 Oo - o0 o EH.r 1 q — b. CD CD Q Cd O C W4 >-p l O rH I & r1 C D 03 S Cr3 A bD H ^ CDbO ~ or>l C 0 ^ '._C rL ~ d c' CD OC H CD ~ ~ D O Cd EH P Q H 0 O H C

r4 oH C T I I I I I I I I I V,0 0 N X N.-4 0 O C d00 00 C 000 IH o II I I I 1 (.. 1 P- H0 0 0 0 0 0 0 0 0 0 0)HO O o 0 J Io < - CO 0 C r — C r —I r —I r-1 CD4 +! I r — 0 L v 1 I 0 0> a) o\ a )r -P0P 0 a 0 8 0 0H C 0 0 0 1 -I -I iI i I I I ( C O O H H,C O ' 0.. 'H 0 - 1 - - - - -t D T — — ) C t r - -iH (J ) Ln ua H P 0 O \ H c r H 0l 00 0 CO O-C Cp- r I rl0 H O 0 0 rd || L — < ( I t' 0L b: C3 a t e l rl l o s n < I4 CQ \ eH co O oo IzdR H HL> HCOO-\ H HkJO '- -; C E C; X! a) L0 0' (CD + - cx3 D 7 0 -p a) L' - 0- o 0 L H - 0 e- 0 b V m (D CO_ _ _ _ _ _ CD DCd(' CDP i HIIIr-dllC r r4H C HH D bh 0 >L ( )C -D m Hrl pi, i i c0 3 L L000 00 LF 0 rL-04 0 i Cr r HHlrCl HH \r JV H^ _ _ _ _ _ _ _ _P

Table IV Present Limits on Flux of Triplets Type Altitude Upper Limit Reference (cm sec sr) Quarks 1/3 e Sea level 17 x 10-10 t (37) 2/3 e Sea level 3.4 x 1010 t (37) Massive Particles Integral or Frac- -10 t* tional Charges 3.2 km 5.0 x 10 Present work "Muonic Quarks" 2/3 e Underground 1.5 x 1010 (22a) (60mwe) 2/3 e Underground -10 (2200mwe) 1.5 x 10 (22b) t 90% confidence level * From last column of Table III

* Supported in part by the National Science Foundation and the U. S. Office of Naval Research contract Nonr 1224(23). tOn leave from Tata Institute of Fundamental Research, Bombay, India. REFERENCES 1o M. Gell-Mann, Phys. Letters 8, 214 (1964). 2. G. Zweig, CERN preprints 8182/TH, 401; 8419/TH, 412 (unpublished). 3. Z. Maki, Progo Theor. Physo 31, 331, 333 (1964). 4. H. Bacry, J. Nuyts and L, Van Hove, Physo Letters 11, 255 (1964); Y. Nambu, Second Coral Gables Conference on Symmetry Principles at High Energy, edited by B. Kursunoglu et al., p. 274 (W. H. Freeman and Co, San Francisco and London, 1965). 5. T. D. Lee, F, GtUrsey and M, Nauenberg, Phys, Rev. 135, B467 (1964). 6. M. Y. Han and Y. Nambu, Phys. Rev. 139, B1006 (1965). 7. W. Blum, S. Brandt, V. T. Cocconi, 0. Czyezwski, J. Danysz, M. Lobes, G. Kellner, D, Miller, D.ReO. Morrison, W. Neale, and Lo G. Rushbrooke, Phys. Rev. Letters 13, 353a (1964). 8, L. B. Leipuner, W, T. Chu, Ro C. Larsen and R. K. Adair Phys. Rev. Letters 12, 423 (1964).

9. V. Hagopian, W, Selove, RP Ehrlich, E, Leboy, R. Lanza, D. Rahm, and M. Webster, Phys. Rev. Letters 13, 280 (1964). 10. A. W. Sunyar, Ao J. Schwarzchild and P. I. Connors, Phys, Rev. 136b, 1157 (1964). 11. T. Bowen, D, A. DeLise, R. M. Kalbach and L. B. Mortata, Phys. Rev. Letters 13, 728 (1964). 12. D. A. DeLise and T. Bowen, Phys. Rev. 140, 458 (1965). 13. T. Massam, Th. Muller and A. Zichichi, Nuovo Cim. 40A, 589 (1965). 14. H. Kasha, L. B. Leipuner, and R. K. Adair, Phys. Rev, 150, 1140 (1965), 15. H. Kasha, L. B. Leipuner, T. P. Wapgler, J. Alspector and R. K. Adair, Phys. Rev. 154, 1273 (1967). 16. R. C. Lamb, R. A. Lundy, T. B. Novey and D. D. Jovanovic, Phys. Rev. Letters 17, 1068 (1966). 17. P. Franzini, B. Leontic, D. Rahm, N, Samios and M. Schwartz, Phys. Rev. Letters 17, 196 (1965). 18. D. E. Dorfan, J. Eades, L. M. Lederman, W. Lee, and C, C. Ting, Phys. Rev. Letters 14, 999 (1965)o 19. G. Damgaard, P. Greider, K. H. Hansen, C. Iverson, E. Lohse, B. Peters and T. Rengarajan, Copenhagen. Private communication. 20, B. K. Chatterjee, G. T. Murthy, S. Naranan, B. V. Sreekantan, M. V. Srinivasa Rao, and S. C. Tonwar, Proceedings of the International Conference on Cosmic Rays, London, 1965.

21. B. Dayton, F. Mills, C. Radmer, L. Wo Jones, Uo Camerini, M. L. Good, and A. Subramanian, MURA Report No. 718 (1966). 22a. J, C. Barton and C. T. Stockel, Phys. Letters 21, 360 (1966). 22bo J. C. Barton, Proc. Phys. Soco 90, 87 (1967). 23. W. A. Chupka, J. P. Schiffer, and C. M. Stevens, Phys. Letters 17, 60 (1966). 24. 0. Sinanaglu, B. Skutnik, and R. Tousey, Phys. Rev. Letters 17, 785 (1966). 25. W. R. Bennett, Jr., Phys. Rev. Letters 17, 1196 (1966). 26. Y. B. Zeldovich, Lo B. Okun, and S. B. Pikelner, Usp. Fize Nauk, 87, 113 (1965). 27. G. Becchi, G. Gallinaro, and G. Morpurgo, Nuovo Cimento 39, 409 (1965). 28. P. Franken and David Rank, University of Michigan. Private communication. 29. G. Domokos and T. Fulton, Phys. Letters 22, 546 (1966). 30. F. Chilton, Do Horn, and R, J. Jabbur, Phys. Letters 22, 91 (1966). 31. G. Damgaard, P. Greider, K. H. Hansen, C. Iverson, E. Lohse, Bo Peters, and T. Rengarajan, Phys. Letters 17, 152 (1965). 32., Yash Pal and S. N. Tandon, Proceedings of the International Conference on Comic Rays (London, 1965), Vol. II, p. 727,

33. Jens Bjorneboe and Zero Koba, Supple Progr, Th. Physo Nos. 37 and 38, 192 (1966). 34. D. E. Lyon, Jr. and A. Subramanian, Design of Ionization Spectrometers Using Iron and Scintillators for the Detection of Hadrons in the 100-1000 BeV range. Submitted to Nuclear Instruments and Methods (1967)o 35. R. K. Adair and N. J. Price, Phys. Revo 142, 844 (1966), 36. See for example W. Galbraith, Extensive Air Showers, Academic Press, 1958, p. 29. 37. R. Gomez, H. Kobrak, A. Moline, J. Mullins, C. Orth, J. VanPutton, and G. Zweig, Phys. Rev. Letters 18, 1022 (1967).

FIGURE CAPTIONS Figure 1. Some possible distributions in the time of arrival of massive particles, with respect to the associated air shower arrival. Figure 2. Experimental arrangement. The air shower counters, not shown, were arranged in 30 ft groups deployed on all four sides of the apparatus, slightly above the spark chamber location. Figure 3. The time of arrival distribution for events whose calculated energy was greater than 100 BeV. This distribution is taken to be the experimental time resolution in the experiment. Figure 4. The time of arrival distribution obtained for all events collected in the experiment. The smooth curve is the experimental time resolution. Figure 5. The distribution in E for events delayed with c respect to the air shower arrival by more than 30 ns. Figure 6. Distributions in n for all events, in various energy bins. Figure 7. Distributions in n for events delayed by more than 30 ns.

Figure 8. Upper limits (99% confidence level) to the cross-section for the production of massive particles (in pairs) in nucleon-nucleon collisions.

0 Xa = 120 gm/cm2 E = 100 BeV; M = 10 BeV Xh =C C, -5 ( \1~~"Xh 85 gm/cm2 -10 0 200 400 600 NANOSECONDS Figure 1

PROPORTIONAL COUNTER PROPORTIONAL COUNTER SPARK CHAMBER K -- 68 gm/cm2 n~\\\\\\\\\\\\\\\\\\\\\\\\\\\i\\\\\l ---- 65 gm/cm2 Id, 4 ux A: Be,:X6:XN,133 gm/cm2 ---- 133 gm/cm2 LI\ ant \":\ *S: 4\T 2 f X 0:.i <:0 4 6 g 133 gm/cm2 133 gm/cm2 CALORIMETER ~400 gm/cm2 1065 gm/cm2 I,,,,,,,,,,,, \ \ \ \ \ \ \ \ \ \,,, \ \ \ \ \ \ \ \ \,, \ \ \ \I IRON K- 2 Feet - I SCINTILLATOR SCALE Figure 2

10,000 7500 U) UL. 0 2 5000 z 2500 -80 -40 0 40 80 NANOSECONDS Figure 3

0 0 z F — w Lu LuD 0 0 z 0 0 CCa anII~~ I - to O 0 0L 8X. 0 O fON 80 0 0 0 0 - C so 0Q_ o — B~~~~~ S1N3A3 d0 t39r —nN

0 ) -0 z w w rfo It Wi I I,, l SI.N3A3 JO 38WBIAllN L^

12,000- 24,000 12,000 22,000 - looooI l 1 1 24,0001 _ 10,000 20,000 0,000 18,000 - (1) (/D (D )Z 0 Z 8000 W > > LtiW 14,000 W L_ LL bL 0 6000 0 12,000 0 6000 Lw I 0 10,000 WL mD C m: 4000- D 8000- > D 4000 < 10 BeV 6000 10-30 BeV 30-50 BeV 2 4000 2000 20_02000 23 4 56 2 3 4 5 6 12 3 4 5 6 NUMBER OF LAYERS > 1 PARTICLE 18,000 - I) (.) 16,000 -F Z W W 14,000 -12,000 -12,000-.LL 0 0 10,000 - 10,000 -Ir 0r W I --- m800 8000 -D D Z 6000- Z 6000 -50-100 BeV > 100 BeV 4000- 4000 -2000 2000 -1 T 4 5 6 2 3 4 5 6 NUMBER OF LAYERS >1 PARTICLE Figure 6

(I) pz L - (D uJ UJ -jn F wU u~w > 0 o I~~~I ~~~0)~~~~~~~~ c'J P o Ag~~i~ r'_ S0 N 3A- -iO 0 3 A nN LL 0 m> - 00 (0) )- O 0SN3A9 d0 <I jfN W _LN:3 A:3 _AIO 0 183 E1,' n N

I000.0 z + 1,, x 120 gm / c FJ L) 10.0 i. +2 O ~ X2gm/cm / h Z - +. I 0 5 lo 15 20 MASS IN Bev Figure 8 U ) Figure 8

Unclassified Security Classification DOCUMENT CONTROL DATA - R&D (Security cllasification of title. body of abstract and indexing annotation must be entered when the overall report in classeiied). ORIGINATING ACTIVITY (Corporate author) 2^. REPORT SECURITY C LASSIFICATION The University of Michigan Unclassified Department of Physics Zb GROUP Ann Arbor, Michigan 3. REPORT TITLE A SEARCH FOR MASSIVE PARTICLES IN COSMIC RAYS 4. DESCRIPTIVE NOTES (Typo of report and inclusive dates) Technical Report No. 29 S. AUTHOR(S) (Last name. first name, initial) Jones, Lawrence W., Lyon, Donald E., Murthy, P. V. Ramana, and DeMeester, G. 6. REPORT DATE 7. TOTAL NO. OF PAGES 7b. NO OF REFS July 1967 37 37 8a. CONTRACT OR GRANT NO. 9a. ORIGINATOR'8 REPORT NUMBER(S) Nonr-1224(23) 03106-29-T b. PROJECT NO. NR-022-274 c. 9 b. OTHER REPORT NO(S) (A ny other numbers that may be assigned this report) d. Technical Report No. 29 10. A V A ILABILITY/LIMITATION NOTICES Distribution of this document is unlimited. 11. SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY Department of the Navy Office of Naval Research Washington, D.C. 13. ABSTRACT A search for heavy particles with a mass greater than that of a nucleon, the existence of which has been suggested by higher symmetry schemes, was performed with an apparatus set up at Echo Lake, Colorado (elevation 10,600 ft). The search was sensitive to strongly interacting particles with masses in the range 5-15 BeV, with no restriction imposed on their electric charge. The method used was to measure the time interval between the arrival of strongly interacting particles and accompanying air shower particles. This information, coupled with information from a measurement of the particle's energy and range of absorption in a total absorption spectrometer, enabled a distinction to be made between massive elementary particles, nucleons, and nuclei. In an operating period of 1542 hours and with an aperture of 0.78 m sr, one delayed event was found whose behavior in the total absorption spectrometer was atypical of a nucleon or nucleus. If one considers this event to represent the arrival of a massive particle then its mass, calculated assuming that its production occurred 1 km above the apparatus, is approximately 6. BeV. This one event corresponds to a flux of the order of 01010 (cm2 sec sr)-l, where a correction for detection efficiency has been included. DD *JAN 1473 Unclassified Security Classification

_ Unclassified W3i11I03 Security Classification 3 s __ 14. LINK A LINK B LINK C KEY WORDS KEY WORDS ROLE WT ROLE WT ROLE WT INSTRUCTIONS 1. ORIGINATING ACTIVITY: Enter the name and address imposed by security classification, using standard statements of the contractor, subcontractor, grantee, Department of De- such as: fense activity or other organization (corporate author) issuing (1) "Qualified requesters may obtain copies of this the report. report from DDC." 2a. REPORT SECURITY CLASSIFICATION: Enter the over- (2) "Foreign announcement and dissemination of this all security classification of the report. Indicate whether DDC is not authorize "Restricted Data" is included. Marking is to be in accordance with appropriate security regulations. (3) "U. S. Government agencies may obtain copies of this report directly from DDC. Other qualified DDC 2b. GROUP: Automatic downgrading is specified in DoD Di- users shall request through rective 5200. 10 and Armed Forces Industrial Manual. Enter the group number. Also, when applicable, show that optional -. markings have been used for Group 3 and Group 4 as author- (4) "U. S. military agencies may obtain copies of this ized. report directly from DDC. Other qualified users 3. REPORT TITLE: Enter the complete report title in all shall request through capital letters. Titles in all cases should be unclassified. If a meaningful title cannot be selected without classification, show title classification in all capitals in parenthesis (5) "All distribution of this report is controlled. Qualimmediately following the title. ified DDC users shall request through 4. DESCRIPTIVE NOTES: If appropriate, enter the type of.,___ report, e.g., interim, progress, summary, annual, or final. If the report has been furnished to the Office of Technical Give the inclusive dates when a specific reporting period is Services, Department of Commerce, for sale to the public, indicovered. cate this fact and enter the price, if known. 5. AUTHOR(S): Enter the name(s) of authOr(s) as shown on 11. SUPPLEMENTARY NOTES: Use for additional explanaor in the report. Enter last name, first name, middle initial, tory notes. If military, show rank and branch of service. The name of the principal author is an absolute minimum requirement. 12. SPONSORING MILITARY ACTIVITY: Enter the name of the departmental project office or laboratory sponsoring (pay 6. REPORT DATE: Enter the date of the report as day, ing for) the research and development. Include address. month, year; or month, year. If more than one date appears on the report, use date of publication. 13. ABSTRACT: Enter an abstract giving a brief and factual summary of the document indicative of the report, even though 7a. TOTAL NUMBER OF PAGES: The total page count it may also appear elsewhere in the body of the technical reshould follow normal pagination procedures, i.e. enter the port. If additional space is required, a continuation sheet shall number of pages containing information be attached. 7b. NUMBER OF REFERENCES: Enter the total number of It is highly desirable that the abstract of classified reports references cited in the report. be unclassified. Each paragraph of the abstract shall end with 8a. CONTRACT OR GRANT NUMBER: If appropriate, enter an indication of the military security classification of the inthe applicable number of the contract or grant under which formation in the paragraph, represented as (TS). (S). (C), or (U) the report was written. There is no limitation on the length of the abstract. How8b, 8c, & 8d. PROJECT NUMBER: Enter the appropriate ever, the suggested length is from 150 to 225 words. military department identification, such as project number, 14. KEY WORDS: Key words are technically meaningful terms subproject number, system numbers, task number, etc. or short phrases that characterize a report and may be used as 9a. ORIGINATOR'S REPORT NUMBER(S): Enter the offi- index entries for cataloging the report. Key words must be cial report number by which the document will be identified selected so that no security classification is required. Identiand controlled by the originating activity. This number must fiers, such as equipment model designation, trade name, military be unique to this report. project code name, geographic location, may be used as key 9b. OTHER REPORT NUMBER(S): If the report has been words but will be followed by an indication of technical conassigned any other repcrt numbers (either by the originator text. The assignment of links, rules, and weights is optional. or by the sponsor), also enter this number(s). 10. AVAILABILITY/LIMITATION NOTICES: Enter any limitations on further dissemination of the report, other than those Unclassified Security Classification