Development of a Rapid Moisture Content Method For …infohouse.p2ric.org/ref/31/30795.pdfCompost Science & Utilization, (1998), Vol. 6, No. 3,14-25 Development of a Rapid Moisture

Compost Science & Utilization, (1998), Vol. 6, No. 3,14-25

Development of a Rapid Moisture Content Method For Compost Materials

Daryl McCartney and Judy Tingley Department of Civil & Geological Engineering, University of Manitoba,

Winnipeg, Manitoba, Canada

One of the most important design and operating parameters at composting facilities is moisture content. Existing standard gravimetric oven methods required a minimum of 19 hours to complete. In this study, by reducing the amount of sample, the time required to complete a standard gravimetric oven method was reduced to nine hours. A new infrared moisture balance method was also developed that reduced the analysis time to 33 minutes. The new infrared method, however, requires one apparatus for each result. The relative precision of the new infrared method was compared to two standard methods using five different materials: straw, celery, paper industry biosolids, municipal biosolids, and cured municipal compost. The new infrared method gave equivalent accuracy on all the materials tested, except for the low moisture content (five percent) material (straw).

Introduction

Moisture content (MC) is one of the critical design and operating parameters used in compost engineering systems. As a result, MC analysis is one of the most commonly performed analytical methods on composts, composting feedstocks, and composting mixtures. In this study, a faster standard gravimetric method and a rapid infrared MC method were developed for compost. The time to complete a standard gravimetric method was reduced from 19 to nine hours, and the new infrared method required only 33 minutes to complete. These faster methods could be used during feedstock preparations, during process monitoring, and for final product testing.

In this study, we: 1) investigated the precision of a standard oven-drying method using smaller sample amounts; 2) developed a new infrared MC method; 3) tested the new infrared method for equivalency to two standard oven-drying methods; and 4) compared the analysis time and the total method time requirements of the new infrared method to a standard oven-drying method.

A review of the existing standard methods for MC analysis revealed two MC methods for compost based on gravimetric oven drying (CAN/BNQ 1996; FC- QA0,1994). Additional standard methods were also found for biosolids, peat soil, and fruit (APHA 1995; AOAC 1995a; AOAC 1995b; ASTM 1987; Topp 1993):Table 1). A rapid method for soils using a microwave oven was found, however, it was reported .

that the method was not accurate for samples containing a significant organic fraction (Liu and Evett 1997). Others, (Leonard and Abboud 1997) have reported successful use of a microwave oven method for composting materials. All the standard methods used gravimetric oven drying techniques. Methods 2 and 3 (Table 1) are those recommended in the Canadian compost quality guidelines document (CCMEJ996). Method 1 was more rigorous than the others as it included: 1) detailed descriptions of sample preparations; 2) a water bath step for fluid materials (high MC materials); 3) the narrowest drying temperature range (103 to 105°C); and 4) it specified that du- plicate analyses should agree within five percent of their average. The increased

14 compost Sciences Utiliition Summer 7998

M1 NI

3 . . -

2..

3..

4

5

6

rii bl

tlt a

f u dl tii

fa P' h//

Ul

P'

fct Sii

an a dl all P' fii T ail

b),

Sii

CG

ClC

Development of a Rapid Moisture Content Method for Compost Materials

TABLE 1. Comparison of the various standards for moisture content

Material Sample Sample Drying Method Developed Size Depth Drying Temp. Drying Method No. For fcm) Method ("C) Time No. Reference

1. Biosolids 25 to 50 NS' Oven 1045 1 Ovemight, until weight change < 4% or 50 mg, whichever is less

2. Peat soil 10 to 12 NS' Oven 105 16h

3. Peat soil 50 3 Oven 105+-5 16 h,oruntil no change in mass of sample

4. Soil NS' A few O\en 105 t 5 24 to 48 h, or until weight change <0.1","

5. Compost 150 ds' Thin layer Oven 105 Until constant weight (-24 h)

6. Fruits 20 NS' Oven 703 2 h intervalsf

2540 G APHA, 1995

AOAC, 1995a

ASTM, 1987

Topp. 1993

FCQAQ, 1994

AOAC,

1 NS - not specified 2ds- grams of dry solids 3 70°C at a pressure Oi less than 100 mm Hg (13.3 kPa) "2 hour intervals until consecutive weirhinrrs varv 5 3 mr

rigour was probably due to the paste-like properties and the variability of municipal biosolids. Method 6 was distinct from the other methods. This method recognized that many fruits contain organics that are volatile at 105"C, therefore the method uses a lower temperature (70°C) under a vacuum to drive-off the water from the sample.

The accurate measurement of compost MC, bulk density, carbon, nitrogen, sul- fur, specific heat content, etc. is important for estimating handling, storage, treatment, disposal, and reuse costs for composting (US EPA, 1997; Terashima et al. 1984; Mar- tin et al. 1995). The overall accuracy depends on the sampling and analysis techniques used. Excellent discussions of general sampling protocols for organic materials are found in Kratochvil et al. (1984) and Kratochvil and Taylor (1981). Specific sampling protocols for solid waste materials are available in ASTM (1993); US EPA, 1997; and Martin et a/. (1995).

Martin et a!. (1995), developed, implemented, and evaluated a detailed sampling protocol for quantifying the components in a municipal solid waste stream. During sampling, they found putrescibles (food waste and the materials contaminated with food waste) to be the most heterogenous (variable) component within a mixed waste stream. This high sampiing variabiiity, however, should not carry over into the MC analysis. As Terashima et a1.(1984) reported, the MC analysis was the least affected by a reduction in sample size. These researchers evaluated four parameters: MC; bulk density; combustibles; and ash. It is important to note that the majority of heavy met- als would be associated with the ash. Upon incrementally reducing an original sample size from 150 to 4.7 kg (1/32 reduction), they found a relative error of less than five percent for MC, while that for incombustibles (ash) was greater than 70 percent. They also observed that the use of stratified random sampling increased the MC analysis accuracy by 30 percent, while no improvement was observed for incombustibles (ash). Stratified random sampling requires one to segregate a sample verti- cally and horizontally before random samples are taken, This technique is of particu-

967.03 A Method I

D 2974-87 Method A

51.2

11 I .

920.151

Compost Science 8 Utilization Summer1098

Da yl McCartney and Judy Tingley

lar importance for samples that exceed their field capacity, i.e., all the moisture in the sample is not physically held by the solids, allowing some to move off the solids.

To better understand the impact of MC value accuracy on conceptual design calculations, we conducted a sensitivity analysis. Typical conceptual design calculations must estimate the volume of material to be composted. This determines: 1) the size of material handling and processing equipment, such as shredders and screens; and 2) the size of composting reactors, e.g., the area of the composting pad for windrow systems or the size of the building for in-channel systems. The volume es- timation equations used were those developed by Rynk et al. (1993). These equations were based on mass balance relationships with primary consideration given to MC and the carbon to nitrogen (C:N) ratio. To assess the impact of a change in MC on the *

volume calculations the following approach was used: 1) selecting an initial MC target of 55 percent and using the design equations presented in Rynk et al . (1993) the volume of material to be composted (V,) was calculated; 2) for each design scenario the MC of one of the feedstock materials was then decreased by 5 percent and the volume of material was then recalculated (V,); 3) the material volume change (AV,%) resulting from the five percent increase in MC was then calculated using equation 1; 4) steps 2 and 3 were repeated for each feedstock material; and 5) steps 1,2,3, and 4 were repeated for each design scenario.

The value of five percent was selected because both APHA (1995) and Terashima ef al. (1984) suggested a relative analytical error of five percent was satisfactory for MC Getermina tions.

AV = ~ ~ 1 * 1 0 0 . . . . . . . . . . . . . . . . . . . . . . . . . .

Four conceptual design scenarios were analyzed, each of which had been com- pleted at the University of Manitoba for actuaI clients. Typical values for MC, nitrogen, C:N ratio, and bulk density were used for the feedstock materials investigated (Rynk et al. 1993). The analysis did not consider changes in bulk density with changes in MC, Le., the only value changed when calculating V, was the MC. The results of the analysis are summarized in Table 2. Of these scenarios, the worst- case (wood chips used as an amendment for food waste) would result in a 22 percent relative error in the volume estimates. For conceptual design purposes, a range of 3.20

percent was considered ade- TABLE 2. quate, therefore we were

comfortable using five per- A summary of the sensitivity of feedstock mixture voIumes to a 5?40 change in the MC value of the wet cent as the target percent rel- or dry materials used in various composting scenarios

ative s tandard deviation Volume Change ResultingFrom (%RSD) for our precision

-_________ In summary, several stan- One Fish waste 80 15.5 dard methods were found in

Materials A 5% Change In assessments. ‘ 7 Feedstock Scenario Material MC (“10) Material MC (‘Ye)

Saw dust 39 Two Food waste 60

Wood chips 40 Three Yardwaste 38

7.5 the literature, but all required at least 16 hours of oven-drying. Although there is a large

11.5 22.0 0.5

Water 100 NA sampling error (high variabil- Four Municipal biosolids 80 13.0 ity) associated with sampling

Straw 12 1 .o putrescibles in the municipal NA - Not Applicable. solid waste stream, the ana-

1Yl C E

att M1

tl- sii

ru

tw le 6 re

W!

Vii

IT ti!( frr

tl- 3::

bii

Oil

dj ~

trr bi, b,,

R: ta tc su b;; i r

SB

U:’

S II

Summer 1998 Cii

Devrlopinent of a Rapid Moisture Content Method for Compost Materials

lytical error associated with the MC determination is low. To achieve the low analytical variability, however, sample size reduction and blending must be thorough. A relative analytical error of less than five percent was identified as a satisfactory value for MC determinations.

Materials and Methods

The APHA (1995) protocols for method development and evaluation were used in this study. First the method procedures were developed and the single-analyst precision of the method was determined. A second analyst was then given the method and without any training was required to analyze three unknown samples. The method ruggedness, i.e., the stability of the result produced when steps in the method were varied, was then conducted. The new method was also checked for equivalency with two standard methods. Standard Methods 1 and 6 (Table 1) were selected for equivalency testing. Method 1 will hereafter be referred to as the APHA Method, and Method 6 will hereafter be referred to as the AOAC Method. The new infrared method will be referred to as the IR Method.

Sa rriplcs

The materials sampled were selected to represent a wide range of MC values. The materials were either compost or composting feedstock materials. For the determination of the drying time and precision of the APHA Method, samples were obtained from twenty-two different types of organic materials. Five different materials, repre- sentative of a range of typical materials and MCs, were selected for the comparison of the IR Method to the APHA and AOAC Methods: 1) straw; 2) paper industry biosolids; 3) cured compost; 4) municipal biosolids; and 5) celery. The cured compost was also used for the bias investigation. The straw was year old wheat straw. The municipal biosolids were obtained from the North End Water Pollution Control Centre in the City of Winnipeg and consisted o f combined primary and secondary biosolids that had un- dergone anaerobic digestion and were dewatered using centrifuges. The paper industry biosolids were obtained from the Pine Falls Paper Company in Pine Falls, Manito- ba and consisted of combined primary, secondary, and deinking (recycle) plant biosolids that were dewatered using a belt press. The celery was purchased at a local supermarket. The compost came from the Clear Lake Golf Course facility, located in Riding Mountain National Park, Manitoba, and was produced from a feedstock containing kitchen waste, grass clippings, and straw. Three different samples were used to investigate the affect of the IR Method lamp intensity setting: 1) straw as already described; 2) the cured compost previously described except that, in addition, it was blended with sand; and 3) backyard compost supplied by one of the authors, consist- ing of kitchen waste, leaves, and straw.

Sa ritple Prepnra f ion

Moisture determination was conducted as soon as possible after samples were re- ceived. If the samples were not analyzed immediately, they were stored at 4°C. A rep- resentative sample was placed on a square rubber sheet and the sample quantity was reduced by quartering and placed in a zip lock plastic bag. The analysts worked rapid- ly to prevent moisture loss. Clods within a sample were broken up while still in the plastic bag, either by hand or by tapping lightly with a rubber mallet. Coarse compost

SummerlBgE 17 Compost Science & Utilization

Daryl McCartney and Judy Tingley

feedstock samples, such as, peels or other solid fruits or vegetables, were size-reduced before analysis using a food processor, blender, grater, scissors or knife. As this is a subjective process, the analyst decided which method reduced the size of the sample into more or less consistent size pieces and which was also expedient to minimize moisture loss. The sample was quickly processed and retumed to a tightly sealed con- tainer and refrigerated. Paste like samples, such as dewatered biosolids, were difficult to work with as they tended to stick to implements. A cork borer or a disposable sy- ringe with the lower tip cut off were used to transfer these samples.

APHA Method: Arnericnn Public Health Association Method 2540 G. Total, Fixed, and Volatile Solids in Solid and Serrrisolid Samples

For this investigation, the minimum sample size was reduced from 25 g to 5 g. The sample was placed in a porcelain crucible and dried at 104 t 1°C. The required drying time, determined by repeated two hour drying cycles, was found to be 55 hours. The drying time used in the equivalency study was six hours. A balance accurate to the nearest 0.01g was used. The precision of the APHA Method was also determined using straw samples and various sample sizes (5,25,10, and 50 g).

AOAC Method: Association of Oficial Analytical Chenrists Metliod 920.151. Solids (Total) in Fruits and Fruit Prodticts

This method is similar to the APHA Method, except a vacuum oven set a t a temperature of 70 k 2°C under a pressure of 5100 mm Hg (13.3 kPa) was used to dry the material. The required drying time, determined by repeated two hour drying cycles, was found to be 26 hours. The drying time used in the equivalency study was eight hours. A balance accurate to the nearest 0.01 g was used.

IR Method: Infrared Mechanical Moisttire Balmce Method

Moisture was determined by drying the compost feedstock or compost sample under an infrared bulb. The sample was situated directly on a gravimetric balance such that a direct reading of percent moisture was given once the sample weight sta- bilized. A description of the IR Method is provided in Table 3.

The equivalency of the IR Method was tested against two standard methods: the APHA Method and the AOAC Method. Ten replicates of straw, municipal biosolids, and celery were run using all three methods. Paper mill biosolids and compost were also run using the IR and APHA Methods. The relative analytical precision of the APHA cind AOAC hilethods were compared to that of the IR Method. The bias of the IR Method was compared to that of the APHA Method. Samples for the bias investigation were prepared by oven-drying compost samples overnight, then adding water to achieve calculated MC values of 5,15,45,60, and 75 percent. Each analysis was replicated three times.

The affect of the IR Method lamp intensity setting on drying times and MC values was also investigated. Lamp intensity settings of 80,90, and 100 percent were as- sessed on backyard compost samples. Settings of 70, 80, and 90 percent were as- sessed on a compost that was cured, screened, and blended with sand. Settings of 30,40,50, and 60 percent were investigated on straw. The MC determinations were replicated four times at each setting.

.

Summer 1998 18 compost Science 8 Utilization

the: re¶l

Rall mal wet

- App-’ a. Cb gran b. D)

Prwc

a. Ph b. SW c. D) witR into) moii d. L. e. Si f. Ai! Rectr g. Dl h. ?I

Rep?

a. R: b. R<

__

COf4


Time Requirements

The times required to complete the MC analyses using the modified APHA and the IR Methods were recorded. Both the total time required to produce a result and the required analyst (staff) time were determined.

Statistical Analysis

The data was analyzed using Sigma Stat statistical software (Jandel Scientific, San Rafael, California). One way ANOVA analyses were done using Tukey tests on normal data and Kruskal-Wallis tests on data that was not normal. Bar and whisker plots were prepared for the method equivalency testing (Fox et a1.1995).

TABLE 3. A description of the IR Method protocols

Apparatus

a Cenco Mtxhanical Moisture Balmce equipped with '1 125 watt infrared lamp controlled by a 0 to 100% rheostat and a 5 gram torsion wire b Disposable aluminum pans

Procedure

a Prepare the IR moisture balance according to the m.inutactiirers instructions. b Set a clean disposable aluminum pan on the retainer pnn of the IR balance. c Distribute the sample evenly around the disposcihle pan ancl in sufficient quantity to bring the moveable pointer in line with the fixed index and the O'%. line This is approximately a S gram simple Note that sample material niwt not spill into the retainer p.in under the disposable pan as thi5 material would be included in the weight measurement, but not the moisture content d Lower the lamp housing and turn on the IR lamp e Set the rheostat between 60 to 1W%, depending on the cvpected MC of the material f As moisture is lost from the sample, follow the n imuk~t i i re rs instructions until no further weight change occurs Record the reading, which is the '% moisture k>ss in the sJmple. g Discard the dned sample h The disposable pans may be reused as long as proper balance calibration is performed before each analysis

Reporting

a Record the moisture content to the nearest 0 1%

b Record the rheostat setting used

_ _ -

Restilts

The precision of the APHA Method using tlle shorter dryig time (six hours) and several types of material are summarized in Table 4. The majority of the precision results were within our target %RSD of 15 percent. The three exceptions were both straw samples and the wood chips. These exceptions all had one thing in common. The original samples were heterogenous materials that resisted blending efforts. Although not attempted in this investigation, additional blending or grinding may reduce the variability.

To determine if the precision could be improved for the straw, the use of larger straw samples was investigated using the APHA Method. The resulting straw precision values, mean MC of four samples (%RSD), were 7.25 percent (5.78 percent), 6.78 (11.60 percent), 8.11 percent (6.37 percent), and 8.31 percent (10.53 percent) for the weight groups 50,25,10, and 5 g, respectively. The overall mean MC value for all 16

Compost Science 8 Utilization Summer1998 19

Daryl McCartney and Iudy Tingley

TABLE 4. Moisture Content (MC) of the organic materials evaluated in the study. Values were

determined using the APHA method, where n = the number of samples; x = the mean value; s = the sample standard deviation; and %RSD = the percent relative standard deviation

n S - - x (Yo) - Yo RSD

3 4 0 0 35 8 75 IO 3 3 0 35 10 91 2 8 5 0 57 6 71

Organic Material Sampled

Straw (blended)

Wood chips Compost' cured, screened & blended with sand Composti cured & unscreened Municipal Biosolids2

Cut to S lcm length and mixed

3 25 6 0 51 199 10 51 6 0 44 0 85

2 50 5 1 77 7 51 2 51 5 OM 1 24

With ceramics With leaves With straw With woodchips

3 52 3 2 5 4 78 3 52 1 1 45 2 78

I O 67 1 1 37 2 04

2 74 1 0 07 0 95

6 78 7 0 61 0 78 3 $3 4 0 15 0 18 3 87 1 0 56 Oh4

2 89 3 0 07 0 78 Tomato (blended) 96 1 0 11 0 IO

3 52 6 1 77 3 36

3 72 4 0 1 0 0 14

Backyard compost feedstock niiuture7 Paper industry biosolids' Anaerobic compost feed5tchck mixture5 Orange peel

Backyard compost feedstock in IY titreh

Apple (blended) Broccoli (blended) Fresh leaves (blended)

Biosohds7 Celery (blended) Lettuce (blended)

' Produced from kitchen waste, grass clippings, and straw Municipal wastewater primary and secondary biosolids that were anaerobically digested and centr~fugally dewatered Kitchen waste, freshly fallen Iraves, and shredded straw Paper industry wastewater hiim>lids from primary and secondary processes that are combined wlth deinking plant

biosolids and belt press dewatered On a dry weight basis the materials were 54% office white piper, 18% old newsprint, 17% glass clippings, and 1 1%

kitchen waste As per 3, except no straw added ' Municipal wastewater primary and wcondary biosolids obtained from clarifier under flow

Biosolidsz 3 74 4 1 47 1 98

Potato peel

3 88 0 1 09 1 24

6 96 2 0.12 0 13 IO 96 6 0 15 0 I6 3 96 7 0 26 0 27

- - - _

samples was 7.61 percent with a RSD of 11.59 percent. These results suggest that there was no correlation between the APHA Method precision and the sample size for straw. Samples of the straw were also analyzed using the IR Method at the same time as the sixteen APHA samples were run. Three replicates were run. The mean MC value was 7.97 percent with a %RSD of 0.07 percent. The IR Method, therefore, was more precise than the APHA Method, however the IR Method gave slightly higher results. These results suggested that straw heterogeneity was not the main factor affecting the accuracy of the MC determinations.

The reason for the higher values attributed to the IR Method and the lower accuracy of the APHA Method may be related to the hydrophilic nature of the straw. It was observed that the MC of the straw was related to the ambient relative humidity. For example, the MC values of the straw reported in Table 4 (3.3 and 4.0 percent) were sig- nificantly lower than those reported in the previous paragraph. Both samples were taken from the same straw bale located in our laboratory. The first results (Table 4) were obtained during a dry season, while those reported in the previous paragraph were obtained during a more humid period. Furthermore, the low bulk density of the straw

20 Gmpost Science a Utitization Summer 1998

(me of si

sityi 6.42 face Sam

dry! cur

w 0 1 '

alen The wit: des wi t m u the:: sho) are' O d s j meti umi

indt fere

NO,

achi s e n

Mean SSD' I % R s Rang:

1 Sam 2 Rek

tc tl- 3(( 6.. 7.. tlt

P' ecc dl

rii F;

S ii

UL

C((

e MI

MI

2. 4 MI

C

b tii t n

41

%

2:

f(( 1" tii

ts in in

ru a1 tll 111

MI

Ul

m

0)

S

m

i: (:: M1

C:

Develoament of n Rnnid Moisture Content Method for Comzlost Materials

to lower values, for example, the straw mean MC value at a 30 percent lamp intensity was 6.3 percent as compared to 7.61 percent observed using the APHA Method (from 16 samples used for the straw precision investigation reported in this paper). The longest drying time observed, 33 minutes, was for the wettest material tested (MC=78 percent; Figure 4).

Time Reqiiirement

The tasks required to complete the APHA Method, ex c 1 u d in g preprocessing , were: 1. determine the initial weight and place in oven; 2. dry; 3. remove from oven; 4. cool; 5. determine final weight samples and clean crucibles; and 6. calculate MC values. The relative time requirements for the tasks involving the analyst were 1 > 5 >> 6 > 3. The required staff time was 4.0 minutes per sample (36 samples; '%RSD=27). Reducing the minimum sample size from 25 to 5 grams, decreased the overall analysis time required for the APHA Method from 19 to nine hours. The drying time was reduced from 16 to s ix hours. The IR Method tasks, excluding preprocessing, were: 1. calibrate and zero instrument; 2. load instrument; 3. dry sample; 4. record readings; 5. clean pan. The relative time requirements for the tasks involving the analyst were: 2 >> 4 > 2 = 5. The required technician time was 3.6 minutes per sample (34 samples; SD=?.4). The

Compost Science & Utilization

a,

v)

2 -

1 -

0 :

3 4 .

2 3.; Lamp Intensity Setting

s -c 30 -c 40 - 50 - 60

-

0 5 10 15 20 25 30 35 40

Drying Time (minutes)

Figure 2 Effect o f the infrared Lamp intensity on the moisture content deter- niin.ition for straw simples. Each point represents the mean of four replicates. 'The errof bars represent one standard deviation above and below the mean.

Lamp Intensity Setting +90 -c 80 s -c 70

0 10 15 20


O L - , ' ' 5 " ' ' : ' ' ' ' : ' ' ' ' :'

i:ipirc 0: Effect of thts infrared lamp intensity on the moisture content determi- ii'ition for cured compost samples. Each point represents the mean of four replic.1tt.s. lhe error bars represent one standard deviation above and below the nieaii.

6 I . I I

2oF 10

t i Lamp Intensity Setting

-c 100 -c 90

it

' i

I - 80

' " : " " : " " : ' " ' I " ' : " " : " " : " " : ' ~ ~ . : ~ 0 5 10 15 20 25 30 35 40 45


Figure 4. Effect of the infrared lamp lntensity on the moisture content determination for fresh feedstock misture of backyard compost samples Each point

deviation above and below the mean.

i t ;I represents the mean of four replicates The error bars represent one standard

Daryl McCartney and Judy Tingley

overall analysis time varied (18 to 33 minutes) depending on the type of material. The one significant disadvantage of the IR Method is that one apparatus was required for each measurement.

Summa y & Conclusions

A review of standard MC methods found two methods developed for compost. Both of these methods used gravimetric oven-drying determinations, requiring long drying times (typically overnight) due to the larger sample sizes specified. Based on the work of Terashima et a). (1984) smaller sample sizes were investigated for a gravimetric oven-drying method (APHA Method; APWA 1995). It was found that the time to complete the APHA Method was reduced from 19 to nine hours, while maintaining a satisfactory precision (‘YoRSD 2 5 percent). The APHA Method precision associated with straw samples was higher (%RSD=10 percent) and the precision did not improve when the sample size was increased from 5 to 50 grams. It was concluded that anoth- er factor other than sample heterogeneity must be affecting the straw MC results.

A rapid IR Method was developed and was compared to the APHA Method. The IR and APHA Methods were found to be equivalent using samples with MC values >51 percent and not equivalent using samples with very low MC values (MC<5 percent). Equivalency testing was not conducted on samples with MC values between five and 51 percent. It is recommended that careful attention be given to the lamp intensity setting of the IR apparatus. It was found that a low setting resulted in a method bias and extended drying times. This was most evident when working with samples containing a low MC. The IR Method yielded a MC result much faster than the APHA Method, however, one apparatus was required for each IR Method result. The required analyst time was comparable for the two methods.

In conclusion, a faster gravimetric oven-drying method and a rapid infrared method were developed and tested. These new methods would interest those who need MC values in a hurry.

Acknowledgement

We would like to acknowledge the contribution of Hector Poggi for the idea of using the 113 Method for compost moisture content determination.

References

AOAC. 1995a. Chapter 2: Fertilizers. I n : Official Methods of Analysis of AOAC International, 16th Ed., Vol. 1: Agricultural Chemicals; Contaminants; Drugs. Association of Official An- alytical Chemists International, Arlington, Virginia, USA. pp. 35-36.

AOAC. 1Y95b. Chapter 37: Fruits and Fruit Products. I n : Official Methods of Analysis of AOAC InternationaI, 16th Ed., Vol. 1: Agricultural Chemicals; Contaminants; Drugs. As- sociation of Official Analytical Chemists International, Arlington, Virginia, USA. pp. 5.

APHA. 1995. Method 2540 G. Total, fixed, and volatile solids in solid and semi-solid samples. In: Standard Methods for the Examination of Water and Wastewater, 19th Ed. American Public Health Association, Washington, DC, USA. pp. 2-58.

ASTM. 1987. Standard Test Methods for Moisture, Ash, and Organic Matter of Peat and 0th- er Organic Soils (D 2974-87). American Society for Testing and MateriaIs, Philadelphia, PA, USA.

CAN/BNQ. 1996. Moisture content. In: Organic Soil Conditions - Compost. Standard number CAN/BNQ 0413-200. Bureau de Normalization du Quebec, 70 Dalhousie St., Suite 220, Qu&bbec, Quebec, G1K 4B2, Canada. pp. 10.

.

24 - p a t Science a Uttliiatlon Summer 1998

CC

F C

Fo)

KXT

KIT

L e

Li.

M1


CCME. 1996. Guidelines for Compost Quality. Canadian Council of Ministers of the Environ- ment Documents, Manitoba Statutory Publications, 200 Vaughan St., Winnipeg, Manito- ba, R3C 1T5, Canada.

FCQAO. 1994. Water Content. In: Methods Book for the Analysis of Compost. Federal Com- post Quality Assurance Organization. Cologne, Germany. pp. 9-1 1.

Fox, E., K. Shotton and C.Ulrich. 1995. SigmaStat Statistical Software User’s Manual. Jandel Corporation. San Rafael, California.

Kratochvil, B., D. Wallace and J.K. Taylor. 1984. Sampling for chemical analysis. And. Cliem.

Kratochvil, B. and J.K. Taylor. 1981. Sampling for chemical analysis. Anolyticnl Chemistry. V.53:

Leonard, J. and S. Abboud. 1997. Personal Communication. University of Alberta and the Al- berta Research Council, respectively. Edmonton, Alberta, Canada.

Liu, C. and J.B. Evett. 1997. Determining the moisture content of soil (microwave oven method). I n : Soil properties: Testing, measurement, and evaluation. 3rd Ed. Prentice Hall.

Martin, J.H., A.R. Collins and R.G. Diener, R.G. 1995. A sampling protocol for composting, recycling, and re-use of municipal solid waste. Air & Waste Management Assoc. V.452364- 870.

Rynk, R. et nl. 1992. On-farm Composting Handbook. Published by the Northeast Regional Agricultural Engineering Service. Ithaca, NY.

Terashima, Y., S. Urabe and K. Yoshikawa. 1984. Optimum Sampling of Municipal Solid Wastes. Conservation & Recycling, 7:295-308.

Topp, G.C. 1993. Soil water content. I n : Soil Sampling and Methods of Analysis. M.R. Carter, Ed. The Canadian Society of Soil Science. Lewis Publishers. Ann Arbor. pp. 541-557.

US-EPA. 1997. Test methods for evaluating solid waste: Physical/chemical methods. United States Environmental Protection Agency. EPA SW-846.

1984. V. 56~113R-129R.

924 A-938A.

pp. 35-40.

Summert9m 25 Compost Science 8 Utilization