ArticlePDF Available

Abstract and Figures

Th e objectives of this research were to (i) test a new oxygen barrier (OB) stretch fi lm with a 20-fold lower oxygen permeability than the polyethylene (PE) fi lm commonly used on farms to wrap bales and (ii) determine the eff ects on microbial status, dry matter (DM) losses, and fermentation of alfalfa (Medicago sativa L.) bale silage. Five fi eld trials were conducted on a farm near Turin, Italy. In Trial 1, the bales were wrapped with two, four, six, or eight layers of either conventional PE or OB fi lm. A further four trials were conducted to compare bales wrapped with four layers of OB with bales wrapped with six layers of PE. In the outer layer of bales, the pH was lower in the OB than the PE silages. Th ere was a signifi cant eff ect of the fi lm type and number of layers on the percentage of bale surface covered by mold (P < 0.001), with less than 15% in bales wrapped with at least four layers of OB fi lm. Storage DM losses ranged from 50 to 123 g kg−1 DM and were aff ected by the type of fi lm used (P < 0.001) and by the number of layers applied (P = 0.035), with consistently lower values in OB silages. Th e new stretch fi lm supports the possibility of conserving high quality alfalfa silage for up to 14 mo, with less fi lm than currently used.
Content may be subject to copyright.
Alfalfa
942 Agronomy Journal Volume 100, Issue 4 2008
Published in Agron. J. 100:942–948 (2008).
doi:10.2134/agronj2007.0258
Copyright © 2008 by the American Society of Agronomy,
677 South Segoe Road, Madison, WI 53711. All rights
reserved. No part of this periodical may be reproduced
or transmitted in any form or by any means, electronic
or mechanical, including photocopying, recording, or
any information storage and retrieval system, without
permission in writing from the publisher.
B
  has advantages over traditional
hay production, such as a more  exible harvest date, less
weather dependency, and greater  exibility in ration formula-
tion (Savoie and Jofriet, 2003).  e bale silage technique is
characterized by its unique individual-package storage sys-
tem, but is particularly prone to spoilage, because each ton of
baled silage has six to eight times the surface area in contact
with plastic  lm compared to clamp silage and about half of
the silage is within 120 mm of the  lm cover (Forristall and
O’Kiely, 2005). Air penetration in the silage, which stimulates
aerobic bacteria, yeasts, and molds, is the main cause of aero-
bic deterioration; this results in DM and nutrient losses, the
accumulation of pathogens and mycotoxins (Scudamore and
Livesey, 1998), and reduced DM intake.  erefore, to ensure
a good and stable silage conservation, air-tightness has to be
maintained throughout extended conservation periods (Paillat
and Gaillard, 2001).
Traditionally, manufacturers of plastic  lms for bale wrap-
ping tended to use PE because of its mechanical characteristics
and low cost. Typical plastic  lms for stretch-wrap silage are
coextruded, linear low-density PE that are 25 µm thick before
being stretched 50% or more during application (Lingvall,
1995). e low density of 0.92 g cm
−3
causes these PE  lms to
be relatively permeable to oxygen and other gases (McNally et
al., 2005). Oxygen  ow through a 25-µm-thick  lm of a low-
density PE at 0.1 MPa overpressure, 23°C, and 85% relative
humidity is 7120 cm
3
m
−2
d
−1
(American Society for Testing
and Materials, 1980). Another important factor that is o en
neglected is the in uence of temperature on the permeability
coe cient that increases exponentially with temperature.
is leads to a 7.4-fold increase in permeability when  lm
heats to 70°C in the sun (Daponte, 1994). Daponte (1992)
also reported that the oxygen permeabilities of PE silage  lms
from low quality to high quality are close to each other in
the range of the thickness because a nearly linear relationship
exists between permeability and  lm thickness. McNally et al.
(2005) showed that PE  lms with low density have poorer gas
barrier properties than materials of higher density, crystallin-
ity, and orientation as a result of extrusion processing condi-
tions. Paillat and Gaillard (2001) reported that stretching to
60% reduced the thickness from 25 to 19 µm, accelerated  lm
wear, and decreased the service life of the  lm by 48% on aver-
age. Furthermore, Hancock and Collins (2006) reported that
the oxygen permeability of a single layer of PE  lm stretched
to 150% of its original length increased to values ranging from
7750 to 9810 cm
3
m
−2
d
−1
, varying according to the manufac-
turing process.
Increasing the number of  lm layers markedly improves air-
tightness because of the increase in distance the air has to travel
to reach the forage. Four layers of PE are usually applied in two
subsequent and complete rotations of the bale with an overlap
of 50% between the layers (Savoie and Jofriet, 2003). A signi -
cant reduction in mold growth and an improvement in silage
conservation quality were obtained when six or eight layers of
lm were applied compared to four (Keller et al., 1998; Müller,
2005).  is is particularly true when high DM forages, and
especially alfalfa, are ensiled in wrapped bales and conserved
for periods more than 8 mo. More layers of stretch  lm assure
a better airtight cover but involve prohibitive increases in costs,
ABSTRACT
e objectives of this research were to (i) test a new oxygen barrier (OB) stretch  lm with a 20-fold lower oxygen permeability
than the polyethylene (PE)  lm commonly used on farms to wrap bales and (ii) determine the e ects on microbial status, dry
matter (DM) losses, and fermentation of alfalfa (Medicago sativa L.) bale silage. Five  eld trials were conducted on a farm near
Turin, Italy. In Trial 1, the bales were wrapped with two, four, six, or eight layers of either conventional PE or OB  lm. A further
four trials were conducted to compare bales wrapped with four layers of OB with bales wrapped with six layers of PE. In the outer
layer of bales, the pH was lower in the OB than the PE silages.  ere was a signi cant e ect of the  lm type and number of lay-
ers on the percentage of bale surface covered by mold (P < 0.001), with less than 15% in bales wrapped with at least four layers of
OB  lm. Storage DM losses ranged from 50 to 123 g kg
−1
DM and were a ected by the type of  lm used (P < 0.001) and by the
number of layers applied (P = 0.035), with consistently lower values in OB silages. e new stretch  lm supports the possibility
of conserving high quality alfalfa silage for up to 14 mo, with less  lm than currently used.
New Oxygen Barrier Stretch Film Enhances Quality of Alfalfa
Wrapped Silage
Giorgio Borreani and Ernesto Tabacco*
Dip. di Agronomia, Selvicoltura e Gestione del Territorio, Univ. of Turin, via
Leonardo da Vinci, 44, 10095 Grugliasco (TO), Italy. Received 25 July 2007.
*Corresponding author (ernesto.tabacco@unito.it).
Abbreviations: cfu, colony forming units, CP, crude protein; DM, dry matter;
F, type of stretch lm e ect; HPLC, high performance liquid chromatography;
L, number of layers e ect; NH
3
–N, ammonia nitrogen; PE, polyethylene, OB,
oxygen barrier; TN, total nitrogen.
Agronomy Journal Volume 100, Issue 4 2008 943
in plastic usage, and in environmental concern due to the dis-
posal of the additional plastic (Lingvall, 1995).
e rapid development in wrapping bale technology has led
to a great improvement in the ensiling process, by increasing
bale densities with round balers equipped with a crop-cutter
(Borreani and Tabacco, 2006), reducing working times with
combined baler-wrapper machines (Münster, 2001), and
improving uniformity of plastic distribution on the bale sur-
face with a new-concept wrapping system (Borreani et al.,
2007). However, PE has a permeability to oxygen that is too
high for this kind of application and, therefore, becomes the
weak link in the entire conservation process; a new plastic
material with low gas permeability is, therefore, required and
needs to be tested for baled silage.
e aim of this work was to test the e ects of two 25m
self-sealing stretch  lms (conventional PE and a new OB  lm)
and of an increasing number of plastic layers on the microbial
status, DM losses, and fermentation characteristics of baled
silage made from wilted alfalfa.
MATERIALS AND METHODS
Site and Experimental Design
Five trials were performed at the experimental farm of the
University of Turin in the western Po plain, northern Italy
(44°50´ N, 7°40´ E, altitude 232 m above sea level, annual
mean temperature 11.7°C, and annual average rainfall 739
mm). e research was conducted in 2005 and 2006 on a pro-
ducing 2.0-ha alfalfa  eld at  ve di erent harvests.
e  rst trial (Trial 1) in 2005 was to study the e ects of the
type of stretch  lm (conventional PE and OB) and the num-
ber of plastic layers applied (two, four, six, or eight layers) on
fermentation quality, mold development on bale surface, and
DM losses. Four further trials (Trials 2, 3, 4, and 5) were con-
ducted on a  eld scale to support the hypothesis of reducing the
amount of plastic consumed without reducing the conservation
quality of bale silage, by using four layers of the new OB  lm
instead of six layers of PE  lm.
Trial 1
A second cut of alfalfa was mown with a rubber roll mower-
conditioner at 11:00 h. at a stubble height of 50 to 80 mm
and forage was spread by tedder over the whole  eld surface
within 4 h. A er 1 d of  eld wilting, the forage was raked
and baled ( xed chamber, no knives in the pickup, net wrap-
ping, Columbia R500/Z, WOLAGRI, Suzzara- MN, Italy) in
600-mm-long and 600-mm-diam. round bales and the silage
bales were individually wrapped (FW 500/Z, WOLAGRI,
Suzzara- MN, Italy) using two, four, six, or eight layers of
conventional PE stretch  lm (white low density PE, 250 mm
wide × 25 m thick, Campanini Ugo S.p.A., Pieve di Cento-
BO, Italy, with an oxygen permeability at 1 bar overpressure,
23°C, and 85% relative humidity of 7160 cm
3
m
−2
d
−1
, 50%
stretched) or two, four, six, or eight layers of OB stretch  lm
(transparent coextruded and without UV protection, 250 mm
wide × 25 m thick, speci cally produced for this experiment
by IPM S.p.A., Mondovì- CN, Italy, with an oxygen permeabil-
ity at 1 bar overpressure, 23°C, and 85% relative humidity of
400 cm
3
m
−2
d
−1
, 50% stretched).  ree individual randomly
selected bales were wrapped for each treatment.  e bales were
stored indoors in the dark on their ends in a single tier for over
9 mo (273 d of conservation).
Trials 2 to 5
First, second, third, and fourth cuts of alfalfa were mown
with a rubber roll mower-conditioner at a stubble height of
50 to 80 mm, and forage was spread by tedder over the whole
eld surface within 2 to 4 h. A er 1 to 3 d of  eld wilting, the
forages were raked and baled ( xed chamber, no knives in the
pickup, net wrapping, Columbia R10, WOLAGRI, Suzzara-
MN, Italy) in 1200-mm-long and 1200-mm-diam round bales
and the silage bales were individually wrapped (Kverneland
Silawrap 7525, Kverneland Group, Naerbo, Norway) using six
layers of conventional PE stretch  lm (white low density PE,
500 mm wide × 25 m thick, Campanini Ugo S.p.A., Pieve di
Cento- BO, Italy, 50% stretched) or four layers of OB stretch
lm (transparent coextruded and without UV protection, 500
mm wide × 25 m thick, speci cally produced for this experi-
ment by IPM S.p.A., Mondovì- CN, Italy, 50% stretched). Four
individual randomly selected bales were wrapped for each treat-
ment.  e bales were stored outdoors on a concrete surface on
their side in a single tier for conservation periods of 331, 418,
122, or 167 d for respective trials. All the bales were covered by
a 200-m white plastic  lm to avoid bird damage and to pro-
tect them from direct sunlight.
Forage Sampling and Silage Evaluation
e DM yield was estimated by weighing samples from
four randomly selected 1.8 × 4 m
2
areas at mowing time. Four
herbage subsamples were taken for DM and chemical analyses.
A er baling, two cores were taken from the side of the bale
from a depth of 0 to 480 mm with a core sampler (45 mm
diam.) and mixed together for DM concentration analysis.
A er the conservation period, before removing the plastic  lm,
each bale was examined carefully for visible holes or damage
due to stem punctures. On removal of the plastic  lms, all
visible molds on the bale surface were located and measured,
according to O’Brien et al. (2005).  e percentage of the total
surface area a ected by fungal growth was then calculated for
each bale. Four cores were taken from the side of the bale from
a depth of 0 to 120 mm with a core sampler (45 mm diam.) and
mixed together, for microbiological analyses. Samples were also
taken from a depth of 121 to 480 mm from the bale surface
using the same holes, for microbiological and fermentative
analyses.  e corer was disinfected between samples and bales
using 95% industrial methylated spirits. In addition, the outer
part of the bale (30 mm) was manually removed, weighed,
chopped, and subsampled for microbiological and fermentative
analyses, to better characterize the silage closer to the plastic
lm. e silage was chopped with a laboratory chopper (proto-
type CNR-IMAMOTER, Turin, Italy) set at a 50-mm theo-
retical length.  e chopper knives were disinfected between
bales using 95% industrial methylated spirits.  e samples for
microbiological analysis were immediately stored at 4°C before
analysis later in the day. Furthermore, in Trial 1, the dry weight
losses were evaluated by weighing the bales a er wrapping at 0,
36, 113, and 273 d of conservation.
944 Agronomy Journal Volume 100, Issue 4 2008
Analytical Procedures
e DM concentration was determined on herbage samples
at 90°C dried in a forced-dra oven until constant weight.
Herbage subsamples were dried for chemical analyses by oven
drying to a constant weight at 60°C, then air equilibrated,
weighed, and ground in a Cyclotec mill (Tecator, Herndon,
VA) to pass a 1-mm screen.  e dried samples were analyzed
for total N (TN) by combustion (Micro-N nitrogen analyser,
Elementar, Hanau, Germany) and for crude protein (CP) (TN
× 6.25).  e silage samples were subsampled and immediately
analyzed for the DM concentration by oven drying at 80°C
for 24 h and for microbiological counts. Colony-forming units
(cfu) of yeasts and molds were counted using the pour plate
technique with 40.0 g L
−1
of yeast extract glucose chloram-
phenicol agar (DIFCO, West Molesey, Surrey, UK) a er incu-
bation at 25°C for 3 and 5 d for yeast and mold, respectively.
Clostridial spores were determined following the most prob-
able number technique with lactate-acetate agar (Spoelstra,
1984) a er incubation at 37°C for 7 d.
Wet silage samples (50 g) were homogenized and extracted
for 4 min in a Stomacher blender (Model 400, Seward Ltd,
London, UK) in water (220 mL) or in H
2
SO
4
0.1 N (220
mL).  e pH was determined in the water extracts. Ammonia
nitrogen (NH
3
–N), determined using a speci c electrode,
was quanti ed in the water extracts.  e lactic and volatile
fatty acids (acetic, butyric, propionic acids) were determined
by high performance liquid chromatography (HPLC) (Canale
et al., 1984). Ethanol was determined by HPLC, coupled to
a refractive index detector, on a Aminex HPX-87H column
(Bio-Rad Laboratories, Richmond, CA). Duplicate analyses
were performed for all the determined parameters. Duplicates
were averaged and the means (replicated bales) were considered
as observations in the statistical analysis.
Statistical Analysis
e data were analyzed for their statistical signi cance by
ANOVA, with their signi cance reported at a 0.05 probability
level using the general linear model of the Statistical Package
for Social Science (v 11.5, SPSS Inc., Chicago, IL). In Trial 1,
data were analyzed by ANOVA utilizing the type of stretch
 l m e  ect (F), the number of plastic layers e ect (L), and the F
× L interaction as  xed factors with three replicates. In Trials
2 to 5, signi cant di erences between means were identi ed by
the P values of ANOVA and e ects were considered signi cant
at P < 0.05.  e mold surface coverage data, expressed as a bale
surface percentage, were analyzed as angular transformed val-
ues (arcsine transformation).  e data of DM losses observed
in Trial 1 were regressed with mold surface coverage as the
independent variable.  e MANOVA analysis of covariance
was used to verify the equivalence of the equations for the
lm types.  e regression lines related to each  lm type were
pooled because they were not signi cantly di erent. e data
of mold surface coverage observed in Trials 1 to 5 were aver-
aged across  eld replicates and regressed on the number of days
of conservation as the independent variable.  e determination
coe cient (r
2
), adjusted for degrees of freedom, and RMSE are
reported.
RESULTS AND DISCUSSION
e DM yields, the DM and the CP concentrations at cut-
ting and at baling, the hours of wilting, and the days of conser-
vation are reported in Table 1. Weather was generally favorable
for  eld drying during each trial and no alfalfa was rained
on when drying in the  eld. e favorable drying conditions
allowed the alfalfa to be harvested a er 27, 48, 56, 72, and 32
h of wilting, for the respective trials. Collectively, these tri-
als encompassed a wide range of climatic conditions resulting
in various wilting periods, DM content at baling, and silage
conservation periods.  e DM yields and the DM and CP
concentrations at cutting were typical of the alfalfa in the Po
plain (Tabacco et al., 2002).  e decrease in CP concentration
was due to mechanical losses that occur during wilting and
was in the range of values observed in the same environment
by Borreani and Tabacco (2006) for alfalfa baled at a similar
DM content.  e DM concentration at baling ranged from 459
to 720 g kg
−1
. Concentrating as much DM per bale as possible
was desirable in terms of plastic consumption and costs, bale
weight, and bale number per hectare (Beaulieu et al., 1993).
Ensiling alfalfa with more than 450 g DM kg
−1
, however,
makes silage more prone to spoilage and may result in molding
and heating, which is apparently related to di culties in the
exclusion of oxygen (Han et al., 2006).
Trial 1
e bale weight ranged from 36 to 48 kg bale
−1
, with an aver-
age value of 42 kg.  e bale DM density ranged from 131 to 176
kg m
−3
, with an average value of 155 kg m
−3
.  ese values were
similar to the DM densities of 157 kg m
−3
reported by Beaulieu
et al. (1993) for round bale silage with
greater dimensions (1200 mm diam. and
1200 mm long), and were consistent with
those reported by Huhnke et al. (1997) in
the range of 101 to 231 kg DM m
−3
.
e chemical composition of alfalfa
bale silages a er 273 d of conservation is
given in Table 2.  e forage DM concen-
trations were relatively uniform across
the  eld and no di erences in silage DM
concentration were observed among
treatments at harvesting. In the present
study, the lactic acid concentrations and
ammonia were not a ected by the  lm
type or the number of layers (P > 0.05)
Table 1. Harvest date, main characteristics of alfalfa herbage at cutting and at baling,
bale weight ,and bale DM density of the ensiling trials conducted at Turin, Italy.
Trial
12 345
Cut second rst fourth second third
Cutting date 23 June 2005 11 May 2006 23 Sept. 2005 2 July 2006 16 Aug. 2006
Wilting time, h 27 48 56 72 32
DM yield, Mg ha
–1
3.9 4.8 1.9 3.3 2.9
DM concentration at cutting, g kg
–1
214 202 239 218 207
CP at cutting, g kg
–1
DM 192 206 199 193 195
DM concentration at baling, g kg
–1
642 459 720 554 677
CP at baling, g kg
–1
DM 164 194 156 153 160
Days of conservation, d 273 331 418 122 167
† DM, dry matter; CP, crude protein.
Agronomy Journal Volume 100, Issue 4 2008 945
and were in the range of values reported by Huhnke et al.
(1997) for legume bale silages of similar DM content wrapped
with at least six layers of PE and analyzed a er at least 6 mo
in storage.  e high concentration of DM likely restricted
fermentation and resulted in relatively low concentrations of
all fermentation acids and NH
3
–N.  is was consistent with
observations reported for wilted and severely wilted ( > 600
g DM kg
−1
) alfalfa round bale silages by Han et al. (2006);
these researchers suggested that a high DM concentration
depresses the total amount of fermentation in silages, resulting
in a higher  nal pH and lower concentrations of fermentation
acids, particularly lactic acid.  ere was a signi cant  lm type
e ect on pH and acetic acid for both the surface samples and
the core samples, with lower pH and greater acetic acid concen-
trations in silage wrapped with OB  lm.  e number of layers
in uenced the pH of the surface samples, with values decreas-
ing with increasing amount of plastic applied.  e pH was in
the range of values reported for legume-grass bale silage wilted
to a higher DM concentration than 600 g kg
−1
(Huhnke et
al., 1997), except for the outer layer (0–30 mm) of the silages
wrapped with two layers of PE  lm, where a pH of 6.45 indi-
cates evidence of silage spoilage, as suggested by Huhnke et al.
(1997).  ere was a signi cant  lm type and number of lay-
ers e ect on the ethanol concentration measured in the outer
layer of the bale, with lower values in silage wrapped with the
OB  lm and decreasing values with an increasing number of
applied layers. A greater level of ethanol in the outer layer of the
silages wrapped with the PE  lm indicated a possibly greater
permeation of oxygen through the cover. McDonald et al.
(1991) pointed out that the survival of aerobic microbes, whose
presence promotes maximum degradation, is shown by an
increased silage temperature and a production of compounds
such as ethanol and 2,3 butanediol. For the PE bales, ethanol
is signi cantly higher with four layers and increases dramati-
cally from two to four layers. It can be supposed that in bales
wrapped with two layers of PE, ethanol was partially oxidized
as suggested by Spoelstra et al. (1988) who reported that in
maize silage exposed to air, ethanol was oxidized by acetic acid
bacteria to acetic acid followed by a rapid oxidation of lactic
and acetic acids when ethanol was depleted. As expected, no
detectable butyric acid was present in these high DM silages.
High butyric acid production is normally associated with
undesirable clostridial fermentations that usually occur when
concentrations of DM are lower than 300 g kg
−1
(McDonald et
al., 1991).
Some holes were observed in the plastic  lm for the bales
wrapped with two or four layers but there was no di er-
ence between the types of stretch  lm (Table 3).  e average
numbers of holes were 6 and 1 holes per bale, for the two and
four layers, respectively, and they were due to puncturing of
the  lm by sti alfalfa stems.  ere were signi cant  lm type
and number of layer e ects on the percentage of bale surface
covered by mold (P < 0.001), with values decreasing when
the number of applied layers increased.  e OB  lm greatly
reduced the area of the surface of the bale covered by mold,
Table 3. Number of holes in the plastic, percentage of bale sur-
face covered by visible mold, and DM losses in relation to the
type of stretch lm and number of layers on second cut of alfal-
fa bale silage after 273 d of conservation at Turin, Italy, Trial 1.†
Stretch lm Layers
Bales
with
holes
Holes
per
bale
Bale surface covered
by mold
DM
lossesEnds Side Total
% g kg
–1
PE 2 3 6.3 100.0 82.5 94.2 123
4 1 0.7 86.7 73.3 82.2 95
6 0 0.0 41.2 30.0 37.5 86
8 0 0.0 37.0 35.0 36.4 80
OB 2 3 6.7 25.5 55.8 35.6 75
4 1 1.0 2.5 18.2 7.7 51
6 0 0.0 0.7 5.8 2.4 51
8 0 0.0 0.3 2.3 1.0 50
F (P value)‡ <0.001 <0.001 <0.001 <0.001
L (P value) ‡ <0.001 <0.001 <0.001 0.035
F × L (P value)‡ 0.615 0.002 0.073 0.863
SED§ 0.20 0.11 0.10 19.9
† DM, dry matter; F, type of stretch lm effect; L, effect of the number of layers;
OB, oxygen barrier stretch lm; PE, polyethylene stretch lm.
‡ Effects were considered signi cant at P < 0.05.
§ For mold surface coverage the SED is reported as an angular transformed value.
Table 2. Fermentative pro le in relation to the type of stretch lm and number of layers on second cut of alfalfa bale silage after
273 d of conservation at Turin, Italy, Trial 1.
Stretch lm Layers
Bale surface (0–30 mm) Bale core (121–480 mm)
DM pH
Lactic
acid
Acetic
acid Ethanol NH
3
–N DM pH
Lactic
acid
Acetic
acid Ethanol NH
3
–N
g kg
–1
g kg
–1
DM g kg
–1
TN g kg
–1
DM g kg
–1
TN
PE 2 664 6.45 1.02 0.53 0.91 24.4 643 5.68 1.26 0.60 1.04 38.5
4 614 5.85 4.02 1.38 6.60 22.4 629 5.63 0.70 0.85 5.46 27.9
6 667 5.80 1.12 0.59 4.54 23.0 693 5.68 0.32 0.48 3.23 21.1
8 681 5.74 0.51 0.39 3.30 15.5 670 5.66 0.54 0.60 3.75 19.1
OB 2 577 5.73 3.71 2.16 2.89 23.1 579 5.63 3.25 2.49 3.40 31.4
4 620 5.59 1.57 1.42 3.06 19.3 642 5.56 0.55 0.84 2.82 17.7
6 634 5.58 1.02 1.03 2.12 15.1 603 5.63 0.64 1.00 2.61 25.3
8 645 5.65 1.25 1.40 1.13 23.2 634 5.62 1.44 1.25 1.76 25.6
F (P value)‡ 0.070 0.001 0.719 0.004 0.023 0.580 0.062 0.026 0.218 0.026 0.136 0.656
L (P value) ‡ 0.289 0.019 0.106 0.220 0.015 0.141 0.781 0.130 0.180 0.291 0.056 0.077
F × L (P value) ‡ 0.430 0.189 0.059 0.132 0.028 0.188 0.317 0.114 0.626 0.212 0.007 0.297
SED 47.6 0.48 1.49 0.58 1.50 6.88 49.7 0.08 1.46 0.76 1.13 9.28
† DM, dry matter; F, type of stretch lm effect; L, number of layers effect; NH
3
N, ammonia nitrogen; OB, oxygen barrier stretch lm; PE, polyethylene stretch lm.
‡ Effects were considered signi cant at P < 0.05.
946 Agronomy Journal Volume 100, Issue 4 2008
with lower values than 8% even in the bales wrapped with
only four layers.  e silage wrapped with two layers of OB  lm
presented the same amount of molded surface as bale silages
wrapped with six or eight layers of PE, whereas the surface of
silages wrapped with two or four layers of PE had more than
80% of the surface covered by molds.  e greater percentage of
bale surface covered by mold was re ected by the higher mold
count observed in the outer layer of silage wrapped with the PE
lm in comparison to the OB  lm, independent of the number
of layers applied (Table 4).  e outer 30-mm part of the bale
accounted for about 20% of the total DM stored in the bale
(data not shown), meaning that 20% of the silage wrapped with
two or four layers of PE had to be discarded before feeding to
animals. e di erences were reduced when a deeper layer of
the bale was considered (0 to 120 mm from the surface).  e
mold count was signi cantly higher only in silages wrapped
with two layers of PE, meaning that more than 50% of the DM
stored in the bale was a ected by spoilage. No di erences in
mold count were observed for the silage sampled in the core of
the bale.  e mold counts in the outer 30-mm layer of the bales
wrapped with PE con rmed the values reported by O’Brien et
al. (2007), who found 4.75 log mold cfu g
−1
in bale silages with
intact  lm and wrapped with four layers of plastic, and 5.18
log mold cfu g
−1
in bale silages with visible damage to their PE
wrapping. Our results are consistent with the data reported
by Keller et al. (1998), who found values of 4.0 to 5.5 log mold
cfu g
−1
when less than six layers of PE  lm were used to wrap
alfalfa silages with a DM concentration ranging from 331 to
547 g kg
−1
.
e DM losses at the end of the storage period ranged from
50 to 123 g kg
−1
DM (Table 3) and were a ected by the type
of stretch  lm used (P < 0.001) and by the number of layers
applied (P = 0.035). Hancock and Collins (2006) reported
that DM losses for alfalfa round bale silage at a DM content of
626 g kg
−1
were generally low but highly variable, averaging 63
± 64 g kg
−1
DM, independent of the amount of PE stretch  lm
applied (from two to six layers per bale).  e pattern of weight
losses during the conservation period di ered from the  rst
36 d of conservation (Fig. 1).  e bales wrapped with at least four
layers of OB  lm showed similar trends in weight loss, which were
consistently lower than those observed in silage wrapped with two
layers of OB  lm or in all silages wrapped with the PE  lm. A er
120 d of conservation, only the bales wrapped with two layers of
stretch  lm (either OB or PE) showed limited increases in weight
losses, whereas the other bales did not show any further weight
losses. e regression equation of the pooled data of DM losses (n
= 24) on the percentage of the bale surface covered by mold are
reported in Fig. 2, where a positive relation can be seen between
the surface covered by molds and the  nal DM losses, with an
adjusted coe cient of determination of 0.66 and a RMSE of 17.4.
Table 4. Mold and yeast count, clostridial spores (MPN), and pH in three zones of the bale in relation to the
type of stretch lm and number of layers on second cut of alfalfa bale silage after 273 d of conservation at
Turin, Italy, Trial 1.
Stretch lm Layers
Bale surface (0–30 mm) Bale surface (0–120 mm) Bale core (121–480 mm)
pH Mold Yeast Spores pH Mold Yeast‡ Spores pH Mold Yeast‡ Spores
–log cfu g
–1
MPN g
–1
–log cfu g
–1
MPN g
–1
–log cfu g
–1
MPN g
–1
PE 2 6.45 4.8 3.7 1.2 5.89 4.1 <1.0 1.0 5.68 2.4 <1.0 1.3
4 5.85 4.4 3.7 1.9 5.70 2.8 1.9 1.0 5.63 1.0 <1.0 1.2
6 5.80 4.2 3.7 1.5 5.79 1.8 <1.0 1.2 5.68 2.2 <1.0 1.5
8 5.74 4.4 4.5 2.2 5.70 2.6 3.3 1.3 5.66 1.7 <1.0 1.2
OB 2 5.73 4.2 3.7 1.0 5.53 1.9 <1.0 1.0 5.63 1.9 <1.0 1.6
4 5.59 3.7 3.6 1.5 5.67 2.3 <1.0 1.4 5.56 1.9 <1.0 1.0
6 5.58 3.7 4.1 1.3 5.58 2.0 <1.0 1.5 5.63 2.2 <1.0 1.5
8 5.65 3.4 3.9 1.3 5.58 2.2 <1.0 1.0 5.62 2.5 <1.0 1.2
F (P value)§ 0.001 <0.001 0.595 0.049 <0.001 0.007 0.304 0.026 0.103 0.771
L (P value)§ 0.019 0.003 0.020 0.134 0.650 0.044 0.077 0.130 0.014 0.186
F × L (P value)§ 0.189 0.369 0.069 0.566 0.054 0.021 0.127 0.114 0.028 0.790
SED 0.48 0.89 0.92 0.60 0.17 1.42 0.23 0.08 0.99 0.47
† cfu, colony forming units; F, type of stretch lm effect; L, effect of the number of layers; MPN, most probable number; OB, oxygen barrier
stretch lm; PE, polyethylene stretch lm.
‡ Statistic analysis was not performed.
§ Effects were considered signi cant at P < 0.05.
Fig. 1. Dry weight losses during conservation in relation to
the type of stretch film and number of layers on second cut
of alfalfa bale silage in Trial 1 at Turin, Italy. PE2, PE4, PE6,
and PE8 indicate polyethylene stretch film of two, four, six, or
eight layers, respectively; OB2, OB4, OB6, and OB8 indicate
oxygen barrier stretch film of two, four, six, or eight layers,
respectively. Each symbol corresponds to the average of four
replicates. SEM = 2.70.
Agronomy Journal Volume 100, Issue 4 2008 947
Trials 2 to 5
ese further four trials were performed at a farm scale to
verify the hypothesis of using four layers of OB  lm instead
of six layers of PE  lm to reduce plastic consumption without
increasing the risk of mold damage.
No holes were observed in either plastic  lm a er the con-
servation period. All silages were well fermented and no di er-
ences were found between the fermentative pro les of the core
samples for the two treatments (Table 5). At high DM con-
centration, the fermentation was restricted and resulted in low
concentrations of all the fermentation products.  e NH
3
–N
concentrations were signi cantly di erent in Trial 3, where the
silage wrapped with PE  lm and analyzed a er 418 d of conser-
vation had an NH
3
–N value of 189 g kg
−1
TN in comparison
with the 49 g kg
−1
of the OB silage. When the outer 30-mm
layer of the bale silage was analyzed, signi cant di erences were
found between the two treatments in all the trials, in terms
of pH, mold and yeast counts, and
surface area covered by mold (Fig.
3). e pH was numerically higher
in the PE silages in all the trials
with values signi cantly higher in
Trials 3, 4, and 5. Yeast counts were
signi cantly higher in PE silages
than OB silages in Trials 2, 3, and
4, whereas it was lower in PE silages
in Trial 5. Mold counts were signi -
cantly higher in PE silages than in
OB silages in Trials 2 and 4. Yeast
and mold counts were slightly lower
than values observed in Trial 1, and
values higher than 3 log cfu g
−1
were
only observed in PE silages of Trial
2 and 4.  e surface area of the bale
covered by mold was only higher
than 15% of the total surface in the
PE silages in Trials 2 and 3, which
were conserved for more than 300 d.
is suggested analyzing the pooled
data of the  ve trials by regressing
the surface area covered by mold against the number of days of
conservation (Fig. 4).  e percentage of the surface covered by
mold increased linearly with an increasing number of days of
conservation, both for PE and OB silages, with regressions with
high adjusted coe cient of determination (0.68 and 0.86 for
OB and PE, respectively) and low RMSE.  e increase in surface
covered by mold per day of conservation was signi cantly higher
in PE silages than in OB silages and a er 300 d of storage more
than 30% of the surface of the bales wrapped with six layers of
PE was covered by mold, indicating a high degree of spoilage.  e
surface covered by mold in the OB silages did not exceed 15%
even a er 418 d of conservation.
e reduction in the oxygen permeability obtained with
four layers of the OB  lm can improve the DM recovered with
Table 5. Fermentative pro les of the core (121–480 mm) of
alfalfa bale silages in trials 2 to 5 at Turin, Italy.
Stretch
lm DM pH
Lactic
acid
Acetic
acid
Butyric
acid‡ NH
3
–N
g kg
–1
DM g kg
–1
TN
Trial 2 PE-6 467 5.44 11.2 3.5 0.0 90
OB-4 452 5.28 11.7 4.5 0.0 73
P value§ 0.240 0.106 0.934 0.499 0.112
Trial 3 PE-6 737 5.22 8.8 0.6 0.0 189
OB-4 703 5.37 7.0 0.2 0.0 49
P value§ 0.142 0.351 0.162 0.169 0.007
Trial 4 PE-6 546 5.27 12.1 9.0 0.6 80
OB-4 562 5.34 9.9 4.9 0.0 55
P value§ 0.375 0.247 0.385 0.277 0.652
Trial 5 PE-6 664 5.67 8.2 2.6 0.3 85
OB-4 690 5.79 11.3 4.7 0.0 91
P value§ 0.301 0.246 0.528 0.694 0.309
† DM, dry matter; NH
3
N, ammonia nitrogen; PE- 6, bales wrapped with six lay-
ers of polyethylene stretch lm; OB- 4, bales wrapped with four layers of oxygen
barrier stretch lm.
‡ Statistic analysis was not performed.
§ Effects were considered signi cant at P < 0.05.
Fig. 2. Dry matter (DM) losses at the end of conservation in
relation to bale surface covered by mold on the second cut
of alfalfa bale silage in Trial 1 at Turin, Italy. PE, polyethylene
stretch film; OB, oxygen barrier stretch film. Regression
equation of the pooled data: DM losses = 0.714 surface molds +
49.60; adjusted r
2
= 0.66; RMSE = 17.43.
Fig. 3. The pH, yeast and mold count in the outer 30 mm of the bale, and surface covered by
mold in alfalfa bale silages in Trials 2 to 5 at Turin, Italy. PE, 6 layers, bales wrapped with six
layers of polyethylene stretch film; OB, 4 layers, bales wrapped with four layers of oxygen
barrier stretch film. Effects were considered significant at P < 0.05.
948 Agronomy Journal Volume 100, Issue 4 2008
the wrapped bale system, by reducing the mold development in
the silage layer closest to the  lm and, consequently, reducing the
weight losses during conservation.  e results obtained with four
layers of the OB  lm consistently support the possibility of reduc-
ing the amount of plastic applied per ton of stored DM in severely
wilted forages and of improving the conservation and micro-
biological quality of silages that can only be obtained with at
least six layers of PE  lm, as suggested by Lingvall (1995) and
Keller et al. (1998).
CONCLUSION
e new OB stretch  lm reduced DM losses and mold spoil-
age in high DM alfalfa silages in comparison to the PE  lm that
is typically used commercially.  e data showed that the bale
surface covered by molds can be kept below 15% with four layers
of OB  lm, whereas it reached up to 80% with the same amount
of PE  lm.  is supports the possibility of making good quality
alfalfa wrapped silage with four layers of plastic instead of the six
or even eight layers commonly suggested with PE  lms for longer
conservation periods than 8 mo.  is new tool may solve prob-
lems that have limited the application of wrapping technology
to extremely wilted alfalfa silage without increasing the amount
of plastic applied, which results in increased costs and environ-
mental concerns. Further experiments should be conducted to
investigate the e ects of the OB  lm on di erent forage crops
and DM concentrations at ensiling.
ACKNOWLEDGMENTS
The authors thank Bartolomeo Forzano (Industria Plastica Monregalese
SpA of Mondovì, Italy) for providing the coextruded barrier plastic films
utilized in the experiment. The authors also like to thank Serenella
Piano, Mara Scaiola, and Matteo Maurelli (Dip. Agronomia, Selvicoltura
e Gestione del Territorio, Univ. of Turin, Italy) for the chemical and
microbiological analyses. This work was partially funded by the Regione
Lombardia, Direzione Generale Agricoltura, Project “MARINSIL.” The
authors contributed equally to the work described in this paper. Mention
of trade names is for the benefit of the reader and does not constitute
endorsement by the Univ. of Turin over other products not mentioned.
REFERENCES
American Society for Testing and Materials. 1980. ASTM Standard method
D 398581. In Annual Book of Standards. ASTM, Philadelphia, PA.
Beaulieu, R., J.R. Seoane, P. Savoie, D. Tremblay, G.F. Tremblay, an R.
ériault. 1993. Eff ects of dry-matter content on the nutritive value of
individually wrapped round-bale timothy silage fed to sheep. Can. J.
Anim. Sci. 73:343–354.
Borreani, G., C. Bisaglia, and E. Tabacco. 2007. Eff ects of a new-con-
cept wrapping system on alfalfa round bale silage. Trans. ASABE
50:781–787.
Borreani, G., and E. Tabacco. 2006.  e eff ect of a baler chopping system on
fermentation and losses of wrapped big bales of alfalfa. Agron. J. 98:1–7.
Canale, A., M.E. Valente, and A. Ciotti. 1984. Determination of volatile
carboxylic acids (C
1
–C
5i
) and lactic acid in aqueous acid extracts of
silage by high performance liquid chromatography. J. Sci. Food Agric.
35:11781182.
Daponte, T. 1992. Coextruded fi lms in silage. Plasticulture 96(4):3544.
Daponte, T. 1994. Barrier fi lms for soil fumigation. Plasticulture 102(2):17–24.
Forristall, P.D., and P. O’Kiely. 2005. Update on technologies for producing
and feeding silage. p. 8396. In R.S. Park et al. (ed.) Proc. 14th Int.
Silage Conf., Belfast, Northern Ireland, 3–6 July 2005. Wageningen
Academic Publ., Wageningen,  e Netherlands.
Han, K.J., M. Collins, E.S. Vanzant, and C.T. Dougherty. 2006.
Characteristics of baled silage made from fi rst and second harvest of
wilted and severely wilted forages. Grass Forage Sci. 61:22–31.
Hancock, D.W., and M. Collins. 2006. Forage preservation method in u-
ences alfalfa nutritive value and feeding characteristics. Crop Sci.
46:688694.
Huhnke, R.L., R.E. Muck, and M.E. Payton. 1997. Round bale silage stor-
age losses of ryegrass and legume-grass forages. Appl. Eng. Agric.
13:451457.
Keller, T., H. Nonn, and H. Jeroch. 1998.  e e ect of sealing and of additives
on the fermentation characteristics and mould and yeast counts in stretch
lm wrapped big-bale lucerne silage. Arch. Anim. Nutr. 51:6375.
Lingvall, P. 1995.  e balewrapping handbook. Trioplast, AB,
Smalandsstenar, Sweden.
McDonald, P., A.R. Henderson, and S.J.E. Heron. 1991.  e biochemistry
of silage. 2nd ed. Chalcombe Publ., Bucks, UK.
McNally, G.M., C. La n, P.D. Forristal, P. OKiely, and C.M. Small. 2005.
e e ect of extrusion conditions and material properties on the gas
permeation properties of LDPE/LLDPE silage wrap lms. J. Plast.
Film Sheeting 21:27–37.
Müller, C.E. 2005. Fermentation patterns of small-bale silage and haylage
produced as a feed for horses. Grass Forage Sci. 60:109–118.
Münster, J.M. 2001. Trends in forage harvesting technology. Landtechnik
56:386–387.
O’Brien, M., P. O’Kiely, P.D. Forristal, and H.T. Fuller. 2005. Fungi isolated
from contaminated baled grass silage on farms in the Irish Midlands.
FEMS Microbiol. Lett. 247:131–135.
O’Brien, M., P. O’Kiely, P.D. Forristal, and H.T. Fuller. 2007. Quanti cation
and identi cation of fungal propagules in well-managed baled grass
silage and in normal on-farm produced bales. Anim. Feed Sci. Technol.
132:283–297.
Paillat, J.M., and F. Gaillard. 2001. Air-tightness of wrapped bales and resis-
tance of polythene stretch fi lm under tropical and temperate condi-
tions. J. Agric. Eng. Res. 79:15–22.
Savoie, P., and J.C. Jofriet. 2003. Silage storage. p. 405467. In D.R. Buxton
et al. (ed.) Silage science and technology. Agron. Monogr. 42. ASA,
CSSA, and SSSA, Madison, WI.
Scudamore, K.A., and C.T. Livesey. 1998. Occurrence and signifi cance of myco-
toxins in forage crops and silage: A review. J. Sci. Food Agric. 77:1–17.
Spoelstra, S.F. 1984. Some methods to evaluate the role of clostridia in silage.
Int. Rep. No. 168. IVVO-DLO, Lelystad, the Netherlands.
Spoelstra, S.F., M.G. Courtin, and J.A.C. van Beers. 1988. Acetic acid bac-
teria can initiate aerobic deterioration of whole crop maize silage. J.
Agric. Sci. 111:127–132.
Tabacco, E., G. Borreani, M. Odoardi, and A. Reyneri. 2002. Eff ect of cut-
ting frequency on dry matter yield and quality of lucerne (Medicago
sativa L.) in the Po Valley. Italian J. Agron. 6:27–33.
Fig. 4. Surface covered by mold (%) in relation to days of con-
servation in alfalfa bale silages in Trials 1 to 5 at Turin, Italy.
PE - 6 layers, bales wrapped with six layers of polyethylene
stretch film; OB - 4 layers, bales wrapped with four layers
of oxygen barrier stretch film; regression equations of the
pooled data of the five trials: Surface molds
PE-6
(%) = 0.183
× days of conservation – 20.5; adjusted r
2
= 0.86; RMSE = 8.9.
Surface molds
OB-4
(%) = 0.0442 × days of conservation – 6.36;
adjusted r
2
= 0.68; RMSE = 3.4.
... Any tears on the film surface may disturb the proper conditions established for the fermentation process inside the bale and lead to the aerobic deterioration of silage [10,11]. Note that standard PE films have relatively high permeability to oxygen (over 7000 cm 3 per m 2 per day), which increases further during sunny days [12]. ...
... Although film airtightness can be improved by the use of multiple film layers (4)(5)(6)(7)(8), this approach would generate even higher disposal costs and a stronger negative impact on the environment [10,12]. Oxygen barrier films were proposed as alternatives to standard PE-based products, providing better air impermeability as well as reducing material usage and costs of wrapping [10,12,13]. ...
... Although film airtightness can be improved by the use of multiple film layers (4)(5)(6)(7)(8), this approach would generate even higher disposal costs and a stronger negative impact on the environment [10,12]. Oxygen barrier films were proposed as alternatives to standard PE-based products, providing better air impermeability as well as reducing material usage and costs of wrapping [10,12,13]. ...
Article
Full-text available
A common method of silage production in Europe is based on the use of cylindrical bales wrapped with polyethylene films. In this study, several modifications of composition of these films were tested for their impact on the microorganisms involved in the ensiling process. Different additives, including nanosilver particles and microcellulose, were analyzed upon the first stage of the experiment. In the second stage, the usability of recycled polyethylene as a film component was assessed. The forage value after ensiling was determined during storage, based on analyses of the content of crude fiber, nitrate nitrogen, total protein, sugars, acids (lactic, acetic, butyric and propionic), pH and dry matter. Microbial forage quality was evaluated by analyses of growth of lactic acid bacteria (LAB) compared to the number of undesirable aerobic bacteria, yeasts and molds. Film properties were also characterized. No statistically significant (p < 0.05) differences were shown for the tested film formulae as compared to standard commercial films. In the second experimental stage, an elevated pH and a slightly higher content of acids were observed for the tested films than for the control sample. In addition, for standard PE film supplemented with nanosilver, a higher number of LAB was detected on the inner surface of the film and in the ensiled material.
... The last case corresponds to typical large bales of 1.2 m diameter and height [10,24,25]. However, in agricultural practice typical bales are 1.2 to 1.3 m diameter and height [6,10,19], where diameter is equal [26][27][28][29][30] or almost equal [9,31,32], but not much bigger, than bale height. Other dimensions of large round bales, for example 1.2 m × 1.6 m Ø [33], or 1.2 m × 1.5 m Ø [34] or 1.5 m Ø [35], are used less frequently. ...
... Consequently, four design problems were solved; however, the solutions for equal optimal bale dimensions resulted directly from those for near-equilibrium by substituting 0 (Sections 3.3 and 3.4). It was assumed that the film parameters (width, mechanical parameters , ) are known, i.e., a practically available plastic film can be used, e.g., a commercial PE film used traditionally due to its mechanical properties and low costs [26,27,39]. Also the global number of film layers wrapped on the bale's lateral surface was taken as a given. ...
... For example, for 1.357 m taking 0.05 we obtain near-equal optimal dimensions * 1.22 m and * 1.162 m. The same regularity holds for wider film 0.9 m, for which the solutions to Problem 1 are given in Table 2. (25) and (26). If grows, then for given both the lower and upper bounds for expressed by Equations (25) and (26) increase provided that the inequality form Equation (24) holds; however, these changes are not significant (see Table 1). ...
Article
Full-text available
For the assumed bale volume, its dimensions (diameter, height), minimizing the consumption of the plastic film used for bale wrapping with the combined 3D method, depend on film and wrapping parameters. Incorrect selection of these parameters may result in an optimal bale diameter, which differs significantly from its height, while in agricultural practice bales with diameters equal or almost equal to the height dominate. The aim of the study is to formulate and solve the problem of selecting such dimensions of the bale with a given volume that the film consumption is minimal and, simultaneously, the bale diameter is equal or almost equal to its height. Necessary and sufficient conditions for such equilibria of the optimal bale dimensions are derived in the form of algebraic equations and inequalities. Four problems of the optimal bale dimension design guaranteeing assumed equilibrium of diameter and height are formulated and solved; both free and fixed bale volume are considered. Solutions of these problems are reduced to solving the sets of simple algebraic equations and inequalities with respect to two variables: integer number of film layers and continuous overlap ratio in bottom layers. Algorithms were formulated and examples regarding large bales demonstrate that they can handle the optimal dimensions' equilibria problems.
... Another innovation aspect regards the improvement of the quality of fi lm applications during wrapping (Borreani et al., 2007b;Bisaglia et al., 2011) and the development of stretch fi lms with a high barrier to oxygen (Borreani and Tabacco, 2008). | References: 1 = https://en.wikipedia. ...
... Furthermore, the stretch polyethylene wrapping system has shown some limits, with regards to sealing effi ciency | (Jacobsson et al., 2002), concerning the high permeability to oxygen of stretch fi lms (Borreani and Tabacco, 2008; and the non-uniform distribution of plastic fi lms between the ends and the curved surface of the bale (Borreani et al., 2007b). Th ese problems lead to undesirable air exchanges over the conservation period, and it has been suggested that an increasing number of plastic fi lm layers is required. ...
... Th ese problems lead to undesirable air exchanges over the conservation period, and it has been suggested that an increasing number of plastic fi lm layers is required. A signifi cant reduction in mold growth and an improvement in silage conservation quality are obtained when six or more layers of plastic fi lm are applied instead of four, especially for high DM content baled forages (Keller et al., 1998;Borreani and Tabacco, 2008). Increasing the layers of plastic fi lm contributes to increasing the hygienic quality of the overall silage, as the bale surface covered by mold is reduced, especially when the baled silage is conserved for long conservation periods ( Figure 4). ...
... Layers of Polyethylene Film. Numerous studies (Keller et al., 1998;Hancock and Collins, 2006;Borreani and Tabacco, 2008;Keles et al., 2009;Coblentz et al., 2016c) have evaluated the number of 1-mil-thick (0.025-mm) layers of PE film used to wrap silage bales as at least a portion of the overall experimental structure. Unlike some other management variables, results from these studies are relatively consistent, and form the basis for solid consensus. ...
... Unlike some other management variables, results from these studies are relatively consistent, and form the basis for solid consensus. Three studies (Hancock and Collins, 2006;Borreani and Tabacco, 2008;Keles et al., 2009) evaluated as few as 2 PE film layers for preserving baled silage crops. Each study found some combination of less desirable fermentation products, a higher final pH, and deterioration of nutritive value compared with 4 or more PE film layers, and thereby concluded that 2 layers were insufficient for acceptable fermentation or nutrient preservation (or both). ...
... Coblentz et al. (2016c) reported little difference in silage fermentation or nutritive value when alfalfamixed grass silages were wrapped with 4, 5, or 6 PE film layers, but it was noted that the risk of puncture was noticeably greater along bale edges between the flat-end and rounded-circumferential surfaces of the bale. Similar risks of internal puncture have been noted by Borreani and Tabacco (2008). Given these results, it seems likely that 4 1-mil (0.025-mm) PE film layers will establish acceptable anaerobiosis for fermentation, but this wrapping standard might not be acceptable to silage producers because of risks of internal or external puncture by rigid stems, vermin, or during additional handling and stacking. ...
Article
Although the concept of ensiling large-round or large-square bales dates back to the late 1970s, many refinements have been made to both equipment and management since that time, resulting in much greater acceptance by small or mid-sized dairy or beef producers. This silage preservation technique is attractive to producers for several reasons, but the primary advantage is a reduced risk of weather damage to valuable forage crops compared with preservation as dry hay. Most core principles for making high-quality precision-chopped silages also apply to baled silages; among these, establishing and subsequently maintaining anaerobiosis are priorities. For baled silages, these priorities are critical, in part because recommended moisture concentrations (45 to 55%) are drier, and particle length is much longer. These factors act to restrict the rate and extent of silage fermentation, often resulting in less production of desirable fermentation acids and a greater (less acidic) final pH. Within this context, preservation of baled silages can be improved by applying polyethylene (PE) film wraps promptly, using an appropriate number of PE film layers (6 to 8), selecting a storage site free of sharp objects or other debris, and by monitoring wrapped bales closely for evidence of puncture, particularly by birds or vermin. Under certain conditions, such as those in which the bale moisture of highly buffered forages exceeds the recommended range, the heterogeneous nature of baled silages coupled with a restricted rate and extent of fermentation may increase susceptibility to clostridial activity compared with precision-chopped forages ensiled at comparable moisture concentrations. To date, research evaluating inoculants or other additives designed to improve the fermentation of challenging forages or aerobic stability has been limited, but should not be discontinued. Development of PE film embedded with an oxygen-limiting barrier has yielded positive results in some trials; however, most differences between these novel formulations and reputable commercial PE film have been related to decreases in yeast and mold counts at the surface layer. Related assessments of fermentation or nutritive value determined on a whole-bale basis have been less conclusive. Baled silages can be produced successfully by adhering to straightforward management principles; as such, this form of silage production is likely to remain popular for the foreseeable future.
... There were rather high amounts of mould also in the straw. This could, for example, be due to the straw not being dry enough before being wrapped in plastic, that there were not enough layers of plastic cover (Borreani & Tabacco, 2008) or that there were holes in the plastic (O'Brien et al., 2008). However, we can conclude that the production and storage of wood chip was not optimal in this experiment and should be done differently to avoid detrimental amounts of mould. ...
Article
The aim of this project was to compare the hygienic quality of the bedding, as well as the parasitic load, cleanliness and lying behaviour in heifers kept on either straw or wood chip deep bedding. Fourteen heifers kept in two pens were housed on straw or wood chip bedding for 43 days, after which the bedding material was switched for 43 additional days. Significantly more mould was found in wood chip than in straw and there was a tendency towards less yeast in wood chip. The heifers had a low faecal parasitic load. The results indicate that the wood chip treatment resulted in somewhat cleaner animals. The heifers had a significantly longer total lying time when kept on straw. In conclusion, wood chip deep bedding could be an option as bedding material, but the impact on hygiene of bedding material and lying behaviour needs to be further investigated.
... Oxygen permeability through the plastic film is a crucial factor for maintaining silage quality in the upper layer of the silo when it is perfectly sealed (Bernardes, Nussio, & do Amaral RC, 2012). Stretch-film of six layers in the present study was above a minimum of four layers of stretch-film to achieve suitable anaerobic conditions, due to rapid reduction of oxygen in bale silo (Borreani & Tabacco, 2008;McEniry, Forristal, & O'Kiely, 2011;McEniry, O'Kiely, Clipson, Forristal, & Doyle, 2007). The heat source for immediate lactic acid fermentation during ensiling is just from ambient temperature. ...
Article
Full-text available
In practical of silage production, variations in the fermentation parameters of silage might be attributable to changes in weather data, especially for ambient temperature. The objective of this study was to investigate how the changed temperature of outside weather condition affects distributions of pH and fermentation products in large round bale silage. The bale silages (1.22 m × 1.25 m ø; density, about 168.9 kg dry matter [DM]/m3) were produced by BR6090 Combi from New Holland Agriculture, wrapped with six layers of stretch film, stored at ambient temperature (2–20°C) and constant temperature (10°C; 15°C; 25°C) for 3, 9, 15, 45 and 90 days. Compared with other silages, silage stored at 25°C had higher (p < 0.05) lactic acid content, lower (p < 0.05) pH, butyric acid content and ammonia‐nitrogen (N) ratio of total N. Fluctuant storage temperature limited lactic acid fermentation process with delayed pH decline and continuous butyric production and crude protein (CP) degradation. The differences of pH, lactic acid, propionic acid and ammonia‐N were considerably less evident in the centre of silages stored at different temperatures, and they increased towards the boundaries. The data shows that fluctuant storage temperature had undesirable effect on the fermentation quality of silage, especially for the external part of silo; relatively high and stable storage temperature for initial stage of ensiling process was necessary, but did not promote a positive response to the homogeneous distributions of pH and fermentation products in large round bale silage.
Article
The purpose of this study was to evaluate the production of gases, temperature, fermentation characteristics, chemical composition and aerobic stability of Tanzânia grass haylage wrapped in films of different thicknesses. Gas production and temperature were evaluated on storage days in 4 × 5 factorial scheme (four films: 27, 10, 11 and 13 µm and five storage times: 0, 7, 15, 30 and 60 days). Fermentation characteristics were evaluated in 4 × 4 factorial scheme (four films: 27, 10, 11 and 13 µm and four storage times: 7, 15, 30 and 60 days). Aerobic stability was evaluated in 4 × 3 factorial scheme (four films: 27, 10, 11 and 13 µm and three times of exposure to air: 0, 48 and 96 h). The lowest pH value of the evaluated haylages was found at 60 days. The largest population of lactic acid bacteria and smallest populations of enterobacteria were observed in the haylage wrapped in the film of 13 µm at 60 days. The haylage wrapped in the film of 13 µm had the highest values of dry matter and crude protein. When using Tanzânia grass for haylage making, it is recommended to use the polyvinyl chloride film of 13 µm and storage time of 60 days.
Article
Poor silage fermentation can affect its acceptance by livestock. Alfalfa from 3 field blocks were baled in large round bales (moisture concentration = 591 ± 43.0 g/kg) and then wrapped with 7 layers of plastic either with (EOB) or without (CW) an enhanced oxygen barrier on the day of baling, or 1, 2 or 3 d after baling in order to examine those effects on subsequent intake and digestibility by gestating sheep. Alfalfa was chopped after approximately 5 mo. of fermentation, and then offered for individual ad libitum consumption by 16 gestating ewes [63.5 ± 1.71 kg avg. body weight (BW)] in a 3-period (63-day) digestion study. Silage moisture and nitrogen (N) decreased linearly (P < 0.05), acid-detergent fiber (ADF) increased linearly (P < 0.05), and neutral-detergent fiber (aNDF) increased at a decreasing rate (linear; quadratic P < 0.05) with wrapping delay within EOB but not CW (P ≥ 0.23). Lactic acid expressed as g/kg of total silage acids decreased linearly (P < 0.05) within CW and decreased at an increasing rate (linear; quadratic P < 0.05) within EOB with increasing wrapping delay, likely because of aerobic deterioration. Digestible organic matter intake (DOMI) decreased linearly (P = 0.03) within EOB and quadratically (P = 0.02) within CW (wrap type × wrapping time after baling interaction; P = 0.04), but other intake and digestibility measurements were not affected (P ≥ 0.15) by wrap type or its interaction with wrapping time after baling. Intake of DM and OM (DMI and OMI, g/kg BW) as well as digestible DMI (g/kg BW) responded linearly and quadratically (P ≤ 0.03) with wrapping delay after baling by initially increasing to 1 d after baling, then declining sharply thereafter. Digestibility of aNDF increased linearly (P = 0.04) with wrapping delay which was likely related to reduced DMI. These values align somewhat closely with those for forage quality and fermentation profiles. Therefore, managing alfalfa silage to ensure more desirable fermentation should also result in higher digestible OMI, which will improve the overall energy status of ruminants.
Article
A key aspect of managing baled silages is to quickly achieve and then rigorously maintain anaerobic conditions within the silage mass. The concept of inserting an O2-limiting barrier (OB) into plastic commercial silage wraps has been evaluated previously, yielding mixed or inconclusive results. Our objective for this study was to maximize the challenge to a commercial polyethylene bale wrap, or the identical wrap containing an OB, by using minimal plastic (4 layers), and then extending storage periods as long as 357 d. Forty-eight 1.2 × 1.2-m large-round bales of alfalfa (Medicago sativa L.) and mixed grass forage (66.3 ± 8.66% alfalfa; DM basis) were made at 2 moisture concentrations [47.5 (ideal) or 36.1% (dry)], wrapped with 4 layers of plastic containing an OB or no OB, and then stored for 99, 243, or 357 d. After storage, yeast counts within the 0.15-m deep surface layer were not affected by treatment (mean = 5.85 log10 cfu/g); mold counts could not be analyzed statistically because 26 bales were nondetectable at a 3.00 log10 cfu/g detection limit, but means among detectable counts were numerically similar for OB (4.74 log10 cfu/g) and no OB (4.77 log10 cfu/g). Fermentation characteristics were most affected by initial bale moisture, resulting in a more acidic final pH for ideal compared with dry bales (5.52 vs. 6.00). This was facilitated by greater concentrations of total fermentation acids (3.80 vs. 1.45% of dry matter), lactic acid (2.24 vs. 0.71% of dry matter), and acetic acid (1.07 vs. 0.64% of dry matter) within ideal compared with dry silages. Plastic wrap type had no effect on final concentrations of any fermentation product. During fermentation and storage, we noted greater change in concentrations of fiber components and whole-plant ash within the 0.15-m deep surface layer than in the bale core, and these changes always differed statistically from 0 (no change) based on pre-ensiled baseline concentrations. Overall, concentrations of water-soluble carbohydrates were reduced (mean = 2.3 percentage units) during fermentation and storage, which resulted (indirectly) in increased concentrations of fiber components and crude protein, as well as an overall energy cost of 2.2 percentage units of total digestible nutrient. It remains unclear under what conditions an OB plastic wrap will consistently benefit the fermentation and preservation of baled silages.
Chapter
An overview was made on plastic usage on livestock farms: what kind of polymers and the amount of different types of plastics used on farms. The attention was then focused on the use of plastics in animal feed conservation, from the beginning in the early 1950s of the previous century with the use of polyethylene till the innovative applications with coextruded oxygen barrier films. All features concerning animal feed conservation by ensiling were discussed focusing on key points, advantages and drawbacks of each plastic type and application. The chapter was concluded with perspective on the use of biodegradable bioplastics in view of increasing sustainability of livestock production.
Article
Full-text available
A field trial has been performed to compare a new-concept selective wrapper (3D) based on biaxial rotation of the film applicators with a conventional rotating-table bale wrapper. The objectives were to evaluate the working rates, total plastic consumption, uniformity of plastic distribution on the bale surface, fermentation characteristics, and mold development in alfalfa round-bale silages. The conventional wrapper was set to wrap four (ST4) or six (ST6) layers of film, while the 3D wrapper was programmed to wrap at least four (MODE5) or six (MODE6) layers over the whole bale surface. The conventional wrapper consumed 0.696 and 1.013 kg of plastic per bale for ST4 and ST6, respectively, and was characterized by the number of layers at the ends of the bale, which was five times larger than that at the side. The 3D wrapper utilized 0.862 and 0.976 kg of plastic per bale for MODE5 and MODE6, respectively, and showed a better uniformity in the number of layers applied at the side and ends of the bale, allowing at least seven layers to be applied over the whole bale surface. The 3D system greatly reduced the damage by alfalfa stems to the plastic on the corners of the bale, and it improved the fermentation quality of the alfalfa bale silage over a 180-day conservation period. The 3D system reduced the amount of plastic used by 4% to 15% in comparison to six layers (ST6), which is the optimum amount for the preservation of alfalfa round-bale silage for storage periods longer than 120 days.
Article
Full-text available
The improvement of wrapped big bale silage quality by increasing bale densities should be an important target. The objectives of this research were to evaluate the effects of the use of a cutting system (15 knives spaced 93 mm apart) installed behind the windrow pickup of a fixed chamber round baler. Three field trials were conducted near Turin, Italy (44�509 N, 7�409 E) by ensiling wrapped round bale of first- and second-cutting alfalfa (Medicago sativa L.) at different dry matter (DM) concentrations. The DM and crude protein (CP) losses at baling, forage mean particle length, wet and dry bale weights, bale density, fermentation pattern, and the conservation losses were evaluated. The DM losses at baling ranged from 0.5 to 2.0% and from 0.7 to 4.7% of DM yield at cutting for unchopped and chopped treatments, respectively. The chopping system increased the percentage of stems shorter than 10 cm from 14 to 38% on a DM basis. Chopping increased DM bale density by about 4% in all the trials. After 140 d of conservation the final fermentation quality of silages was not consistently improved by chopping.
Chapter
This chapter describes various storage methods and losses associated with storage. The tower silo is a vertical cylindrical container that can be used to store various feeds. The most common type of tower silo is the top-unloading silo. Top-unloading silos are generally constructed of precast concrete blocks referred to as "staves". Bottom-unloading silos are constructed of cast-in-place concrete or steel such that the ingress of air is limited to minimize spoilage of the fermented feed. There are two basic types of top-unloading equipment: suspended and surface riding. A bunker silo, sometimes referred to as a "horizontal silo", is a paved area surrounded on two or three sides by near-vertical retaining walls usually <6 m high. Bunker silos are less expensive to build than tower silos. The chapter presents few silage fermentation reactions. Silage storage systems are usually designed to conserve high quality feedstuff for ruminants at a minimal cost. © 2003 by the American Society of Agronomy, Inc., Crop Science Society of America, Inc., Soil Science Society of America, Inc.
Article
This work investigates the effect of extrusion processing conditions and co-monomer type on the gas permeation properties of LLDPE films containing polyisobutylene (PIB). The results show improved gas barrier properties with increasing polymer density and increase in film crystallinity and orientation as a result of extrusion processing conditions such as blow up ratios.
Article
Round bales of ryegrass and legume-grass silages were wrapped with stretch film. Initial bale moisture contents ranged from 25 to over 65% (wet basis). After six months in storage, apparent dry matter losses were not affected by initial moisture content. Significant changes were measured in crude protein, acid detergent fiber, and total digestible nutrients for all forages. However, changes in quality were not significantly affected by initial moisture content. Final moisture content of samples bored from 0 to 100 mm (0-4 in.) and 100 to 230 mm (4 to 9 in.) depths ranged from 14.0% to 71.7%. Moisture content had a significant effect on fermentation. Ryegrass silage samples above 50% moisture content averaged 2.5% of DM lactic acid and 0.5% of DM acetic acid while legume-grass silages averaged 1.0% of DM lactic acid and less than 0.4% of DM acetic acid. Of the 312 samples analyzed, 17.9% had pH values above 6.5; a proportion that was relatively constant for samples having moisture contents less than 65%.
Article
Stretch film, used for wrapping bales for silage, reacts very differently in the tropical Réunion Island compared to use in a temperate climate. Tropical climatic conditions have an adverse effect on the physical properties of the polythene film, in particular air-tightness, resistance to radiation and resistance to high temperatures. Modelling air-tightness tests has contributed to a better understanding of the effects of atmospheric conditions on measurements of pressure differences: temperature and air humidity have relatively little effect, while altitude has a major effect. If the measurements are made with a time lapse between wrapping and testing, tests can be used to compare the air-tightness of the bale wrapped with different films. The service life of film exposed on frames and wrapped roll bales provides the information needed to select film that can be stretched and can provide lasting air-tightness. The results contribute to the knowledge base on film that can be stretched under difficult conditions and on a certification methodology.
Article
Thirty-two sheep (30.0 kg avg. BW) were used to study the effects of dry-matter (DM) content at harvest on the nutritive value of timothy grass silages conserved as round bales of high (52.1%, DM50), medium (39.9%, DM40) and low (23.1%, DM25) DM. Chopped grass of 24.4% DM conserved in a horizontal silo (HS) was used as a control. Gross energy, crude protein (CP) and acid detergent fibre (ADF) contents were similar for all silages (P > 0.05). Neutral detergent fiber (NDF) contents decreased as DM of the silages decreased (P < 0.05). Acid detergent lignin and ash contents were highest in HS and lowest in DM40 silages (P < 0.05). Silage pH and water-soluble carbohydrates decreased with decreasing DM of the silages (P < 0.05), while ammonia-N and lactate levels increased (P < 0.05). Acetate concentrations in round bales were lower than in HS silage (P < 0.05). Significant butyrate concentrations were detected only in DM25 silage. Round-bale silages were chopped before feeding and fed ad libitum. DM intake was...