Direct active measurements of movements of lunar dust: Rocket exhausts and natural effects contaminating and cleansing Apollo hardware on the Moon in 1969



[1] Dust is the Number 1 environmental hazard on the Moon, yet its movements and adhesive properties are little understood. Matchbox-sized, 270-gram Dust Detector Experiments (DDEs) measured contrasting effects triggered by rocket exhausts of Lunar Modules (LM) after deployment 17 m and 130 m from Apollo 11 and 12 LMs. Apollo 11 Lunar Seismometer was contaminated, overheated and terminated after 21 days operation. Apollo 12 hardware was splashed with collateral lunar dust during deployment. DDE horizontal solar cell was cleansed of nominally 0.3 mg cm−2 dust by 80% promptly at LM ascent and totally within 7 minutes. A vertical cell facing East was half-cleaned promptly then totally over hundreds of hours. Each cell cooled slightly. For the first time lunar electrostatic adhesive forces on smooth silicon were directly measured by comparison with lunar gravity. Analyses imply this adhesive force weakens as solar angle of incidence decreases. If valid, future lunar astronauts may have greater problems with dust adhesion in the middle half of the day than faced by Apollo missions in early morning. A sunproof shed may provide dust-free working environments on the Moon. Low-cost laboratory tests with DDEs and simulated lunar dust can use DDE benchmark lunar data quickly, optimising theoretical modelling and planning of future lunar expeditions, human and robotic.

1. Introduction

[2] Apollo astronauts on the Moon in early mornings met major difficulties with clinging dust. Operationally dust was one more risk for astronauts to manage personally, with housekeeping help such as brushes. Deployment by astronauts of multidisciplinary Apollo Lunar Surface Experiments Packages (ALSEPs) transmitting to Earth for a nominal two years was a major success. On 21 July 1969, Buzz Aldrin sheltered a minimalised ALSEP and its 47 Kg Passive Seismic Experiment (PSE) behind a large rock, 17 meters from the Lunar Module (LM) (NASA Photo AS11-40-5951). Hitch-hiking on the side of PSE facing LM was a 0.27 kg matchbox-sized Dust Detector Experiment (DDE), (B. J. O'Brien, Proposal for dust detector experiment on all ALSEPs, NASA SC Control 44-006-054, 1966). On 27 August 1969 PSE was terminated after overheating and system failure (NASA SP-214). In October 1969 Apollo 11 Preliminary Science Report (NASA SP-214) incorrectly stated that DDE “showed no appreciable cell degradation caused by dust or debris from LM ascent” [Bates et al., 1969]. By May 1970 agreement was reached, published October 1970, that significant degradation had occurred [O'Brien et al., 1970].

[3] In June 1970 our 22 December 1969 Principal Investigator's 90-Day Report of Apollo 12 DDE measurements was not included in Apollo 12 Preliminary Science Report (NASA SP-235). A Thermal Degradation Samples Experiment on Apollo 14 fortuitously revealed lunar dust effects assumed due to electric fields [Gold, 1971]. No follow-up is known. Apollo 17 deployed a Lunar Ejecta and Meteorites Experiment designed to measure meteoroids and lunar offspring [Berg et al., 1973]. Its data are now being reinterpreted to study movements of lunar dust at sunrise and sunset [Colwell et al., 2007].

[4] Gaier [2005] lists nine problems from lunar dust, from Apollo 12 landing blind to causing health hazards if inhaled. Theoretical modelling of complex lunar environments forecast properties and movements of dust [Stubbs et al., 2007; Colwell et al., 2007; Lee, 1995]. Lunar dust is an easily mobilised component of the lunar regolith. Particles have an average size of 70 μm and can be sharp and angular, increasing mechanical interlocking. About 20% is smaller than 20 μm, more susceptible to electrostatic forces because of larger charge-to-mass ratios [Colwell et al., 2007]. In Apollo expeditions, dust particles were positively charged through intense photoelectric effects excited by energetic solar ultraviolet and X rays. “There is currently no theory for the charge distribution of grains in a regolith” [Colwell et al., 2007]. Adhesion of lunar dust remains a central but unmeasured issue for human activity on the Moon, its causes uncertain. The last astronaut on the Moon summarised the agreed position: “Dust is the Number 1 environmental hazard on the Moon” (H. Schmitt, The Apollo Experience: Problems encountered with lunar dust, paper presented at the Biological Effects of Lunar Dust Workshop, NASA, Sunnyvale, California, 29-31 March 2005).

[5] In 2006 a NASA website ( on Apollo 11 Lunar Dust Detector stated original computer tapes were misplaced before archiving. Our 173 digital 7-track tapes of Apollo 11, 12 and 14 DDE data are the only copies. Lunar-dust literature contains no reference to benchmark Apollo 11 LM contamination [O'Brien et al., 1970]. So we resumed self-funded analyses of personal files and paper charts, many with resolution of 3.6 minutes, plotted in 1969–70. This report is the first outcome. Efforts are underway with SpectrumData to read again 6 million digital measurements with 54-second resolution. A caveat applies to this report that such higher resolution may yield insights not yet articulated.

2. Lunar Dust Detectors

[6] Only Apollo 12 Dust Detector Experiment (DDE) has the invented configuration of 3 identical 2 cm square, n-on-p 10-ohm-cm solar cells shielded by 1.5 mm silicon, facing East (VSCE), West (VSCW) and up (HSC) (Figure 1). Modified DDEs deployed by Apollo 11, 14 and 15 have all 3 solar cells horizontal, with different covers, for radiation measurements [Bates and Fang, 2001]. The 3 solar cell voltages and 3 temperatures of DDE were changed by an analogue to digital converter on the Moon from continuous voltage values to discrete, step-like digital measurements transmitted in ALSEP's telemetry stream. Accuracy of DDE measurements is known from: extensive pre-flight brightness calibrations; temperature dependence calibrations of solar cells whose temperatures were then measured on the Moon; orbital effects where solar brightness varies with Sun-Moon distances; templates of measurements on other Lunar Days; effects of digitisation. Estimates of dust quantities are made from illustrative lunar dust MLS-1. Sunlight transmitted through 0.5 mg cm−2 dust to an underlying solar cell is weakened causing an output voltage 10% less [Lee, 1995; C. M. Katzan et al., The effects of lunar dust accumulation on the performance of photovoltaic arrays, paper presented at the Space Photovoltaic Research and Technology Conference, Lewis Res. Cent., Cleveland, Ohio, 1991].

Figure 1.

Apollo 12 Dust Detector Experiment (DDE) on the Moon. The vertical East-facing 2 cm × 2 cm solar cell (VSCE) is full-face far left. Patches (arrowed) of “collateral” lunar dust were splashed on the Central Station during deployment. Lunar Module (LM) is Southeast, in the direction of the shortest arrow. (Source of base image is NASA AS12-47-6927.)

3. Measurements During Ascent of Apollo 11 Lunar Module (LM)

[7] Apollo 11 DDE was deployed 17 metres from LM, about 30 cm above the lunar surface. Deployment factors are shown in Table 1 for Apollo 11 and 12. Outputs of DDE before and after ascent of Lunar Module (LM) increased slowly (Figure 2) as the sun rose by about 0.5° per hour. At LM ascent outputs abruptly decreased by 17%, 7% and 0%, equivalent to depositions of illustrative dust MLS-1 of 1 mg cm−2, 0.3 mg cm−2 and zero. Cell temperature increased abruptly by 4°C more than expected during the post-ascent loss of telemetry, ultimately reaching peaks of 102° to 105°C near noon [O'Brien et al., 1970]. The cells are sporadically heavily shadowed [Bates et al., 1969]. Further severe constraints on dust analyses include all cells being horizontal, lack of lunar photographs, incomplete calibration and differences between cells. The DDE was tilted about 3 degrees towards the East.

Figure 2.

Outputs of three horizontal solar cells of the Apollo 11 DDE measured every 54 seconds over 10 hours around ascent of the Lunar Module (LM) [after O'Brien et al., 1970].

Table 1. LM Ascent Details wrt DDE
 Apollo 11Apollo 12
LM Distance from DDE17 metres130 metres
LM Azimuth from DDELM was NorthLM was Southeast
LM Ascent CommentsVertical for 10 secondsTilted to West, abeam ALSEP in 10 seconds
Location on MoonMare Tranquillitatis, 0°41′N; 23°26′EOceanus Procellarum, 3°11′S; 23°23′8″W
Time of LM Ascent 19691754 GMT 21 July1426 GMT 20 November

4. Discussion of Apollo 11 Effects

[8] Apollo 11 DDE made the first measurements showing rocket exhausts (P. T. Metzger et al., Cratering and blowing soil by rocket engines during lunar landings, paper presented at the Sixth International Conference on Case Histories in Geotechnical Engineering, Missouri University of Science and Technology, Arlington, Virginia, 2008) caused significant contamination by lunar dust and debris, resultant overheating and early failure of the PSE, the first major scientific experiment put on the Moon by human hands. Such damage was a foreseen possibility, with 17 meter deployment accepted on this first mission in interests of astronaut safety. These benchmark measurements on the Moon appear unused and unreferenced in theoretical modelling of effects of rocket exhausts.

5. Measurements During Ascent of Apollo 12 Lunar Module

[9] Apollo 12 DDE was deployed 130 metres from LM, about 100 cm above the lunar surface. Astronaut activities caused lunar dust we call “collateral dust” to be splashed inadvertently on hardware, photographically visible within 10 cm of DDE (Figure 1). We show below that, in most fortunate serendipity, collateral dust was splashed on DDE.

[10] On Lunar Day 1 (LD1), the horizontal cell (HSC) and the vertical, West-facing solar cell (VSCW) had peak values as expected, before lunar noon (Figure 3) and lunar sunset, long after LM ascent. Each peak matched pre-flight calibration, so both cells were then optically clean (Table 2). However the peak value reached by VSCE on LD1, at solar elevation angle (θ) of 10.9° about 14 hours before LM ascent, was 3.4% less than its pre-flight calibration (Table 2 and Figure 4). By LD2 VSCE was also optically clean (Figure 4 and Table 2). At LM ascent VSCW had only a small output, measuring albedo from the western moonscape, and is therefore not discussed further here.

Figure 3.

Apollo 12 Dust DDE Horizontal Solar Cell (HSC) measurements of apparent brightness over its first lunar day (LD1). An increase by 6% occurred at LM ascent. Detail shows 80% cleansing was prompt and 20% was - in this digital display -about 7 minutes later.

Figure 4.

Comparison of morning measurements by the vertical East-facing solar cell of Apollo 12 DDE on its first two Lunar Days. For improved clarity of this display, raw LD2 data have not been normalised by almost 1% brighter solar illumination on Lunar Day 2 (Table 2).

Table 2. Peak Outputs of Apollo 12 DDE Solar Cells
Solar CellPre-flight Calibration (millivolts)Lunar Day 1 (LD1) Peak (Raw mv)Lunar Day 2 (LD2) Peak (Raw mv)Raw Peak LD2 (% of LD1)Normaliseda Peak LD2 (% of LD1)
  • a

    Lunar values corrected for 80°C temperature and normalised for different distances from the Sun on different days, but not for less-certain lunation effects caused by radiation and dust (O'Brien, manuscript in preparation, 2009).

Horizontal (HSC)142.5142.5 After LM Ascent141.699.4%98.7%
Vertical Faces East (VSCE)137132.3 @ θ = 10.9° Before LM Ascent136.6 @ θ = 10.1° After LM Ascent & Night103.3%102.2%
Vertical Faces West (VSCW)141141.6 After LM Ascent141.599.9%99.2%

[11] Apollo 12 LM was abeam ALSEP within about 10 seconds (A. Bean, personal communication, 2008). At LM ascent, voltage outputs of HSC and VSCE promptly increased by 5% and 1% respectively (Figures 3 and 4). A detailed plot at 3.6 minute intervals, located only for HSC, shows HSC increased by 1% more 7 minutes after LM ascent (Figure 3, detail), becoming optically clean. As the morning of LD1 advanced after LM ascent, VSCE became cleaner, progressively closer to LD2 values at equivalent solar elevation angles (Figure 4 and Table 3). We could not reliably compare such progressive changes in VSCE with theoretical trigonometric calculations because, over many lunations, effects on VSCE after sunrise are complex (B. O'Brien, Direct active measurements of movements of lunar dust: Effects of lunar sunrise in 1969 and 1970, manuscript in preparation, 2009). Consequently a template was developed from raw, uncorrected measurements of VSCE during Lunar Day 2 (LD2) (Figure 4). These were normalised for the positive orbital factor of brighter sunlight on LD2 but not for less-certain negative environmental degradation factors from radiation and/or primeval lunar dust over the lunation LD1 to LD2. Three segments in time of VSCE were then quantified crudely as three stages of a continuous loss of dust (Table 3). Both HSC and VSCE cells cooled by about 0.2°C at LM ascent. During LD1 their peak temperatures were 80.4°C and 82.6°C, some 20°C below Apollo 11 peak DDE temperatures.

Table 3. Three Segments of VSCE Voltage Early Morning LD1-LD2
Time IntervalSolar Elevation Angle θDifference LD2 (Raw)-LD1Difference as % of LD2Segment Name
24 hours before LM Ascent11° < θ < LM ascent at 21°3.5 mv2.6%Pre-LM Ascent Collateral Dust
30 hours after LM AscentLM ascent < θ < 35°2.0 mv1.6%Post-LM Ascent & Remnant Dust Thinning by Natural Causes
After mid-morning Lunar time35° < θ < 60°0.8 mv0.7%Remnant Dust Removed by Natural Causes

[12] This suite of findings suggests that collateral dust on HSC and VSCE was cleansed by LM ascent and subsequent naturally-occurring effects to make all 3 cells optically clean on LD2. Possible alternative explanations were examined. DDE is tilted about 4 degrees towards the East, deduced from sunrise and noon lunar data, all acquired after LM ascent. A tilt to the East caused by LM rocket exhaust could cause the greatest observed effect at LM ascent, 5% increase in HSC output. However, there are 5 items of evidence against such a cause: the initial and closest blast was from the SouthEast; LM-driven tilt to the East would have caused VSCE output to decrease, but it increased; both HSC and VSCE temperatures decreased even though solar input energy increased; HSC output increased further when LM was far distant; VSCE returned to pre-flight calibration values long after. By contrast, all observed effects are consistent with cleansing of collateral dust except possibly the following. We are yet to understand why peak temperatures of VSCE and HSC were 82.8°C and 84.6°C on LD2 after being 80.4°C and 82.6°C respectively on LD1. Analyses are continuing.

6. Discussion of Apollo 12 Effects

[13] The only consistent explanation of Apollo 12 DDE data on LD1 is that collateral dust of 0.3 and 0.2 mg cm−2 (MLS-1) was splashed on HSC and VSCE respectively during deployment (Figure 1) and later cleansed by processes beginning with LM ascent. Cleansing of HSC was by removal of 80% of 0.3 mg cm−2 collateral dust promptly, likely by the direct blast of rocket exhaust gases, and removal of 20% within 7 minutes (Figure 3, detail). The present digital resolution of 3.6 minutes obscures whether the smaller removal was a separate effect or a diminishing part of the prompt effect.

[14] Cleansing of VSCE was more complex. Less than half its 0.2 mg cm−2 collateral dust was blown off promptly at LM ascent. The remainder fell off gradually to a complete loss possibly by mid-morning (Figure 4 and Tables 2 and 3). We suggest lunar gravity caused dust to fall off VSCE as the electrostatic adhesive force grew steadily weaker when the angle of incidence of sunlight on the vertical cell steadily decreased. All dust had gone after lunar night.

[15] Such a conclusion is supported by detailed analyses of prompt effects at LM ascent. With sunlight incident at 26° giving brightness of 60 mw cm−2 on horizontal cell HSC, it was totally cleaned by the rocket exhausts within 7 minutes. However, sunlight incident at 64° gave 120 mw cm−2 on the vertical VSCE and it was cleaned at LM ascent by less than half. Yet VSCE was much more exposed to the direct blast from LM ascent, facing it at 45°, whereas only a thin, circa 3° wedge of HSC was exposed to the initial LM rocket exhausts. Although we cannot exclude more exposure of HSC as LM passed nearly overhead, such an increased exposure would apply also to VSCE.

[16] Therefore we conclude that all collateral dust remaining after LM ascent fell from the smooth vertical silicon surface of VSCE as the morning advanced and sunlight intensity on its vertical surface decreased. We provisionally conclude that the electrostatic adhesive force was matched by lunar gravity when solar elevation angle of about 45° caused brightness of about 100 mw cm−2 (Table 3). This quantification depends on small differences between 40 year old analogue paper charts of VSCE during LD1 and LD2. Higher confidence may come from 54-second digital data, when available. We caution also that DDE measures processes as they occur over only a few centimetres or a few percent of the local Debye length, micro-dimensions not yet considered by models of lunar electric fields, which are very complex on immensely larger scales [Halekas et al., 2008].

[17] Such a model of lunar electrostatic forces, if validated, could be very significant in lunar operations because of its strong variation with solar elevation angle. We are unaware of any search for or analyses of such effects with Apollo. During Apollo the horizontal lunar surface was illuminated by about 40 mw cm−2. The model implies that Apollo problems with adhesion of dust were not primarily caused by electrostatic adhesive forces except for near-vertical surfaces, which of course includes astronauts. One large-scale scheme of managing electrostatic adhesion of lunar dust may be a sunproof shed to exclude energetic photons of sunlight from working environments. Low-energy photons from electric lights could not create adhesive forces on disturbed ambient dust particles.

[18] Apollo 12 DDE proved the importance of collateral dust. Yet, despite collateral dust being clearly visible on many Apollo photographs for 40 years, its significance and properties appear neglected. We find no evidence this palpable factor was considered in analyses of Surveyor 3 samples brought back by Apollo 12 astronauts, so such analyses may require review.

[19] DDE measurements at LM ascent also fill a gap in before-and-after site assessments of surface dust available for mobilisation. Characteristics of lunar material at Apollo 11 and 12 landing sites were initially similar, but greater thrust levels at about 30 feet altitude produced greater erosion rates at the Apollo 12 site [Mission Evaluation Team, 1970, section 6.1.3]. Further evidence of heavy scouring includes total loss of visibility below 40 feet and sandblasting of Surveyor 3 at 180 meters distance (Katzan et al., presented paper, 1991). Lack of contamination of Apollo 12 DDE by LM ascent might be caused in part by its 130 m deployment and in part by winnowing removal of mobile lunar soil from the site by LM descent on the previous day.

7. Conclusions

[20] Direct active measurements by the minimalist, 270-gram Dust Detector Experiment deployed by Apollo 11 and Apollo 12 astronauts in 1969 quantified for the first time movements of lunar dust and their causes in human activities and natural processes. Engineering cultures, addressing design and management, and scientific cultures analysing causes, can now more reliably research and anticipate magnitudes of known and unknown problems with dust to be faced in future lunar missions.

[21] Apollo 11 DDE, deployed 17 m from LM, proved that rocket exhaust caused dust and debris to contaminate the Passive Seismometer, the first scientific instrument deployed by human hands on a celestial body, leading to its severe overheating and premature failure. By contrast, Apollo 12 hardware had collateral dust splashed on it during deployment 130 m from LM. Rocket exhaust and natural causes then removed this dust from smooth silicon vertical and horizontal surfaces during the first lunar day, so that by Lunar Day 2 all 3 solar cells were as optically clean as in pre-launch calibrations. As to effects of rocket exhaust, these DDE data provide a contrasting pair of the first and much-needed quantitative lunar-surface benchmarks for modelling of rocket plume/lunar soil interactions and optimising future lunar missions. Low-cost laboratory experiments with replica DDEs and simulated lunar dust can now quickly extend knowledge and management measures. As to effects of natural causes of movements of lunar dust, Apollo 12 DDE measured actively and directly for the first time collapses of little-understood adhesive forces holding lunar dust to a vertical and a horizontal smooth silicon plate. A simplistic model of electrostatic adhesive forces is suggested whose strength decreases as solar brightness decreases with decreasing angle of incidence on the host surface. A benchmark strength of this adhesive force for this surface is that it counteracts the force of lunar gravity when brightness is at least 100 mw cm−2, 70% of full direct sunlight at an angle of incidence of 45°, at mid-morning or mid-afternoon, sunlight intensity twice that on the lunar surface during Apollo. This model infers that Apollo astronaut problems from clinging dust, at solar elevations much less that 45°, may have been driven by other forces, such as mechanical bonding properties intrinsic to lunar dust, which could explain partial successes in cleaning with Moon Brushes. It follows that on future lunar expeditions, powerful electrostatic adhesion of lunar dust during the middle half of each lunar day could cause greater dust problems than experienced by Apollo astronauts. A sunproof shed on the Moon may provide working conditions free of electrostatically-clinging lunar dust. Attention is drawn to the importance of collateral lunar dust, much photographed yet generally neglected.